Download Recovery of Locomotion After Ventral and Ventrolateral Spinal

Recovery of Locomotion After Ventral and Ventrolateral Spinal Lesions
in the Cat. II. Effects of Noradrenergic and Serotoninergic Drugs
EDNA BRUSTEIN AND SERGE ROSSIGNOL
Centre de recherche en Sciences Neurologiques, Faculté de Médecine, Université de Montréal, Montreal,
Quebec H3C 3J7, Canada
INTRODUCTION
Adult cats subjected to a massive bilateral ventral and ventrolateral spinal lesion at the lowest thoracic level, severely
The costs of publication of this article were defrayed in part by the payment
of page charges. The article must therefore be hereby marked ‘‘advertisement’’
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
damaging ascending and descending pathways, among others,
the reticulospinal and vestibulospinal pathways, can recuperate
voluntary quadrupedal locomotion (Bem et al. 1995; Brustein
and Rossignol 1998; Gorska et al. 1993a,b; Rossignol et al.
1996). However, as described in detail in a preceding paper
(Brustein and Rossignol 1998), the cats suffered from transient
as well as from long-lasting postural and locomotor deficits,
the severity of which depended on the extent of the spinal
lesion. Early after the lesion (early period), none of the cats
could walk or support the weight of their hindquarters, and
they moved around using only the forelimbs. During the recovery period, which lasted from 3 days up to 3 wk depending
on the extent of the spinal lesion, the cats gradually regained
the ability to support weight and walk with their hindlimbs.
However, even when having reached a plateau period during
which there was no further improvement, the quadrupedal
stepping overground and on the treadmill remained wobbly
with poor lateral stability. The coupling phase between the
homolateral fore- and hindlimb was highly variable, and, in the
most lesioned cats, the fore- and the hindlimbs could walk at
different frequencies, which caused continual deviations in the
coupling phase between the two girdles. These locomotor
deficits often resulted in stumbling or falling on one side,
which limited the cats to only a few consecutive steps at a time
and to low treadmill speeds (0.1– 0.35 m/s). These locomotor
deficits were more pronounced during walking uphill or downhill on inclined treadmill (10°), and the amplitude modulation
(AM) of the hindlimb electromyographic (EMG) activity, usually seen when walking uphill, was absent.
The most extensively lesioned cats did not improve their
locomotor skills very much with time and expressed these
deficits even 300 days postlesion. It was suggested that after
functional and anatomic reorganization, the remaining pathways in the dorsolateral funiculi, such as the corticospinal, the
remaining reticulospinal, and possibly propriospinal, were sufficient to provide the necessary drive for initiation of locomotion as well as some postural control. However, their contribution is limited because the cats cannot maintain constant
fore- and hindlimb coupling or adapt to more demanding
situations such as inclined walking.
It is clinically relevant to try and improve the locomotor
capacities in this animal model of incomplete spinal lesion
because the relative percentage of partial versus complete
spinal lesions is increasing in the population of patients (Tator
et al. 1993). Patients with partial spinal lesions exhibit residual
locomotor capacities but suffer from locomotor and postural
deficits such as difficulties in bearing weight, maintaining
0022-3077/99 $5.00 Copyright © 1999 The American Physiological Society
1513
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on June 18, 2017
Brustein, Edna and Serge Rossignol. Recovery of locomotion after
ventral and ventrolateral spinal lesions in the cat. II. Effects of noradrenergic and serotoninergic drugs. J. Neurophysiol. 81: 1513–1530, 1999.
The effects of serotoninergic and noradrenergic drugs (applied intrathecally) on treadmill locomotion were evaluated in two adult cats subjected
to a ventral and ventrolateral spinal lesion (T13). Despite the extensive
spinal lesion, severely damaging important descending pathways such as
the reticulo- and vestibulospinal tracts, both cats recovered quadrupedal
voluntary locomotion. As detailed in a previous paper, the locomotor
recovery occurred in three stages defined as early period, when the
animal could not walk with its hindlimbs, recovery period, when progressive improvement occurred, and plateau period, when a more stable
locomotor performance was observed. At this latter stage, the cats suffered from postural and locomotor deficits, such as poor lateral stability,
irregular stepping of the hindlimbs, and inconsistent homolateral foreand hindlimb coupling. The present study aimed at evaluating the potential of serotoninergic and/or noradrenergic drugs to improve the locomotor abilities in the early and late stages. Both cats were implanted
chronically with an intrathecal cannula and electromyographic (EMG)
electrodes, which allowed determination, under similar recording conditions, of the locomotor performance pre- and postlesion and comparisons
of the effects of different drugs. EMG and kinematic analyses showed
that norepinephrine (NE) injected in early and plateau periods improved
the regularity of the hindlimb stepping and stabilized the interlimb coupling, permitting to maintain constant locomotion for longer periods of
time. Methoxamine, the a1-agonist (tested only at the plateau period), had
similar effects. In contrast, the a2-agonist, clonidine, deteriorated walking. Serotoninergic drugs, such as the neurotransmitter itself, serotonin
(5HT), the precursor 5-hydroxytryptophan (5HTP), and the agonist quipazine improved the locomotion by increasing regularity of the hindlimb
stepping and by increasing the step cycle duration. In contrast, the 5HT1A
agonist 8-hydroxy-dipropylaminotetralin (DPAT) caused foot drag in one
of the cats, resulting in frequent stumbling. Injection of combination of
methoxamine and quipazine resulted in maintained, regular stepping with
smooth movements and good lateral stability. Our results show that the
effects of drugs can be integrated to the residual voluntary locomotion
and improve some of its postural aspects. However, this work shows
clearly that the effects of drugs (such as clonidine) may depend on
whether or not the spinal lesion is complete. In a clinical context, this may
suggest that different classes of drugs could be used in patients with
different types of spinal cord injuries. Possible mechanisms underlying
the effect of noradrenergic and serotoninergic drugs on the locomotion
after partial spinal lesions are discussed.
1514
E. BRUSTEIN AND S. ROSSIGNOL
METHODS
General experimental protocol
Experiments were carried out on two adult cats (EB7 and EB8)
weighing 4.1 and 2.8 kg, respectively, which were among the most
lesioned cats reported in our previous study (Brustein and Rossignol
1998). The cats first were trained to walk on a treadmill at different
speeds (0.2– 0.7 m/s) until they could maintain a regular locomotion
for periods of ;20 min. Then the cats were anesthetized and implanted with EMG electrodes and an intrathecal cannula. Control
experiments (n 5 7–12) were made during a period of 1– 6 wk to
obtain EMG and kinematic values in the intact state. Thereafter the
cats were submitted to the ventral and ventrolateral spinal lesion at the
thoracic level (T13), and their locomotor recovery was followed and
documented daily as well as long after no further recovery could be
observed.
Different adrenergic and serotoninergic drugs were tested, and their
effects on locomotion were followed for several hours and compared
with the performance of the cat immediately before drug application.
At the end of the experimental series, wheat germ agglutininhorseradish peroxidase (WGA-HRP) was injected caudal (L2) to the
site of spinal lesion (at days 340 and 280 postlesion for cats EB7 and
EB8, respectively). Then 3 days later, the cats were perfused and the
spinal site of lesion and the spinal site of WGA-HRP injection as well
as the brain stem and the motor cortices were processed histologically.
All the surgical procedures and experimental protocols were reviewed
and approved by the University’s Ethics Committee.
Surgical procedures
During the first surgery, performed in sterile conditions and under
general anesthesia, a cannula was inserted into the intrathecal space of
the spinal cord and EMG electrodes were implanted in various foreand hindlimbs muscles.
EMG IMPLANTATION.
The surgical procedure for implantation of
EMG electrodes and the function of the implanted muscles were
detailed in the preceding paper (Brustein and Rossignol 1998). The
implanted muscles included, in the hindlimbs: iliopsoas (Ip) and
sartorius (Srt), semitendinosus (St), tibialis anterior (TA), vastus lateralis (VL), and gastrocnemius medialis (GM) and lateralis (GL); in
the forelimbs: cleidobrachialis (ClB) and the lateral head of the triceps
brachii (TriL). Although all muscles were implanted both on the left
(L) and on the right (R) limbs, only the left side of the animal facing
the video camera was used to illustrate the kinematics.
A Teflon cannula (Teflon tube-thinwall,
size 24 gauge) was inserted through an opening made in the atlantooccipital ligament and gradually pushed through the intrathecal space,
so that the tip was located at L4 –5, as measured before the insertion
using external landmarks. The cephalic end of the cannula was inserted into a right-angle plastic port fixed to the skull by dental acrylic,
next to the EMG connectors, and served for administration of drugs or
sterile saline (0.9%). The inlet was capped to reduce contamination
and was opened only at the time of drug or saline application. The
dead space of the cannula was measured before the insertion and was
determined to be ;100 ml. To ensure its patency, the cannula was
flushed with a bolus of 100 ml sterile saline solution on nonexperimental days.
INTRATHECAL CANNULA.
VENTRAL AND VENTROLATERAL SPINAL LESION.
The ventral and
ventrolateral spinal lesions were performed in a second surgical
intervention, under general anesthesia, when the recordings of the
control period were completed (see Brustein and Rossignol 1998).
Recordings and data analysis
EMG activity and the video images used for kinematic analyses
were recorded simultaneously. They were synchronized using a
SMPTE time code (time code generator Skotel TCG-80N and time
code reader TCR-80N), recorded on both analog and video tapes (for
details, see Brustein and Rossignol 1998).
In addition to quantitative kinematic analysis, the video tapes were
reviewed frequently to obtain a better overall impression of the drug
effects because the kinematic analyses in one plane gives only a
partial assessment of the locomotor improvement. In some cases, the
cats also were filmed walking from above to better evaluate the body
orientation during walking and its improvement after drug injections
(not illustrated).
Drug application
The tested drugs are listed in Table 1 together with their main
action and the given doses. All the drugs were administered to both
cats and were generally repeated in at least three separate experiments
at different days after the partial spinal lesion (also indicated in the
rightmost columns in Table 1).
Before drug application, a walking sequence on the treadmill was
recorded to determine the predrug baseline performance, which varied
with time after the lesion (see Brustein and Rossignol 1998). Then a
bolus of 100 ml of the drug solution, prepared with sterile saline
(0.9%), was administered through the intrathecal cannula using an
adapted syringe to fit the inlet. The syringe was fastened to the rostral
end of the cannula to avoid leakage and to ensure that the whole dose
would be injected. Immediately afterward, a second bolus of 100 ml
sterile saline was injected and served to flush the drug out of the dead
space of the cannula. Usually one dose of one drug was given, unless
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on June 18, 2017
balance, and adapting their walking speed (Barbeau and Rossignol 1994). Therefore the purpose of this study was to try and
improve the residual voluntary locomotor function by applying
different noradrenergic (NE) and serotoninergic (5HT) drugs
through a chronic intrathecal cannula. These two groups of
drugs were chosen because of their known involvement in
initiation and/or modulation of locomotion in complete spinal
cats (Barbeau and Rossignol 1994; Barbeau et al. 1993; Chau
et al. 1998a,b).
The specific purpose of this study was then to investigate the
effects of noradrenergic and serotoninergic drugs on the treadmill locomotion of two cats subjected to extensive but partial
spinal lesions of the ventral and ventrolateral quadrants. We
wanted first to try and express locomotion in the early period
after the lesion when there is practically no hindlimb locomotion and second to improve the locomotor capacities (walking
maintenance and regularity, coupling between the fore- and the
hindlimbs and speed adaptation) at a later period when voluntary quadrupedal walking was reestablished but is deficient.
Therefore, the different drugs were applied in two cats
through an intrathecal cannula with its tip ending caudal to the
spinal site of lesion. The locomotor activity was compared
before and after the drug application using chronically implanted electrodes to identify changes in EMG activity under
constant recording conditions in the same cat and to follow the
progressive evolution of its locomotor capacities during the
recovery period up to the point of achieving a stable locomotor
behavior (plateau period).
Our results show that some drugs (noradrenaline, methoxamine, and quipazine) improved the walking capacities of the
cats, resulting in a more stable and maintained locomotor
performance. However, other drugs, such as clonidine, caused
a deterioration of the locomotor performance. The mechanisms
for the action of the drugs in partial spinal cats are discussed
with reference to clinical situations.
EFFECTS OF NORADRENERGIC AND SEROTONINERGIC DRUGS AFTER PARTIAL SPINAL LESIONS
TABLE
1.
1515
Drugs injected
Dose (mM/100 ml)
Drug Name
Role
Cat EB7
Cat EB8
RBI
0.8–4.7
(9, 14, 17, 34, 49, 104, 107, 147, 182)
0.6–5.6
(16, 33, 128, 169, 190, 205, 210)
2–4
(196, 198, 203, 239)
12.8–20.5
(205, 210, 212)
2.4–7.1
(42, 44, 120, 126, 175, 177)
0.6–2.2
(224, 227, 231, 234)
1.5
(266, 268)
1.1–2.3
(280, 282, 287, 289)
1.6–6.3
(6, 8, 21, 23, 70, 135)
1.9–5.6
(14, 65, 76, 84, 86, 91, 98)
4–6
(104, 107)
12.8–20.5
(84, 86, 91, 98, 100)
1.2–2.3
(16, 35, 63)
1.2–3.4
(112, 119, 127)
1.5
(154, 156)
2.3–4.5
(168, 170, 175)
Norepinephrine
Neurotransmitter
Clonidine
a2-norepinephrine agonist
Sigma
Methoxamine
a1-norepinephrine agonist
RBI
Yohimbine
a2-norepinephrine antagonist
RBI
Serotonin
Neurotransmitter
RBI
Quipazine
5HT agonist
RBI
DPAT
5HT1A agonist
RBI
5HTP
5HT precursor
Sigma
Drugs used in the different experiments are listed according to their role and commercial source. The range of doses tested in each cat is given together with
the days post spinal lesion (in parentheses) on which the experiments were carried out. The early period experiments are marked in italics.
indicated otherwise, and consecutive drug experiments were separated
by $48 h. The doses were kept as small as possible to minimize
central effects or any other discomfort to the cat. The reaction time to
the application of a drug, its maximal effect, and the fading time
depended on the drugs. To cover the whole time course of the effects,
recordings started as soon as 20 –30 min after the drug application and
were repeated each 30 – 45 min until the drug effects wore off (between 2 and 5 h). The results presented illustrate the maximal effects
of the drug on any given day. The drugs were applied during the early,
recovery, and plateau periods postlesion (see Table 1); however, the
illustrated experiments were recorded in the early period when the
cats had pronounced locomotor deficits and during the plateau period
when the cats exhibited consistent locomotor deficits. It should be
emphasized that the duration of the different periods and the severity
of the deficits depended on the extent of the spinal lesion (see Brustein
and Rossignol 1998). For example, for the most severely lesioned cat
(EB7), the early period lasted almost 4 wk, whereas it lasted only 1 wk
for cat EB8.
Evaluation of the extent of the spinal lesion
Evaluation of the extent of the spinal lesion was done by combining
conventional histological staining methods (cresyl violet, KluverBarrera) and by retrograde WGA-HRP labeling. The histological
procedures were detailed in Jiang and Drew (1996), Matsuyama and
Drew (1997), and also in the companion paper (Brustein and Rossignol 1998).
RESULTS
Extent of the spinal lesion and recovery of locomotion
The results of the histological evaluation of the spinal lesions of cats EB7 and EB8 were described in detail in the
companion paper (Brustein and Rossignol 1998) (a summary is
provided in Fig. 1). Briefly, examination of the site of spinal
lesion of cat EB7 showed a major disruption of both ventral
and ventrolateral funiculi. Intact tissue could be found only in
the dorsal columns and in the left dorsolateral funiculus (DLF)
(see Fig. 1A for the photomicrograph of the site of spinal
lesion). The damage to the spinal quadrants was correlated
with a pronounced decrease in the numbers of HRP-labeled
neurons (expressed as percentage of the intact, Fig. 1A) found
in brain stem nuclei corresponding to the origin of reticulo- and
the vestibulospinal pathways descending in these quadrants
(see Table 1 in Brustein and Rossignol 1998, for the number of
HRP-labeled cells). The lesion of cat EB8 (Fig. 1B) encompassed the lateral vestibulospinal pathway, bilaterally, whereas
FIG. 1. Summary of the histological findings related to the extent of the spinal lesion.
Photomicrographs (A and B, cats EB7 and EB8,
respectively) of cross-sections taken from the
spinal site of the lesion (stained with KluverBarrera) illustrate the extent of the maximal
spinal lesion. Graph illustrates the number of
horseradish peroxidase (HRP)-labeled cells
found after the spinal lesion on the left and right
pontine reticular formation (PRF), medullary
reticular formation (MRF), lateral vestibular
nucleus (LVN), red nucleus (RN), and motor
cortices, expressed as percent of the mean values obtained from 3 intact cats. Shaded vertical
bars crossing the 100 level illustrate the coefficient of variation of the average cell counts in
the intact cats. Numbers of HRP-labeled cells
can be found in the companion paper (Brustein
and Rossignol 1998).
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on June 18, 2017
Source
1516
E. BRUSTEIN AND S. ROSSIGNOL
the reticulospinal pathways were damaged but to a lesser extent
compared with cat EB7. In both cats, a major increase was
observed in the number of HRP-labeled cells in the motor
cortex contralateral to the less affected DLF (Fig. 1, A and B),
as well as by changes in the distribution of the cells (Brustein
et al. 1997).
Despite the extensive lesion to the ventral and ventrolateral
spinal quadrants damaging important reticulo- and vestibulospinal pathways, both cats recovered voluntary quadrupedal
locomotion. Yet they suffered from long-lasting locomotor
deficits, as illustrated in Fig. 2. From the consecutive stick
figure diagrams and the angular excursion traces, it is noted
that even at ;120 days postlesion (see Fig. 2, B and D), the
hindlimb stepping was highly irregular compared with the
intact situation (Fig. 2, A and C). The cats also suffered from
poor lateral stability and sudden stumbling, which are reflected
in abrupt amplitude increases of the EMGs bursts (see LSrt of
cat EB7 and RSrt for cat EB8). In addition, the homolateral
fore- and hindlimb coupling was affected as illustrated for the
most lesioned cat, EB7, in Fig. 4A and in the phase plots of Fig.
8B. The coupling phase between the fore-and hindlimb shifted
gradually, a result of different walking rhythms in the fore-and
the hindlimbs. For example, in Fig. 2B, six bursts were counted
in RTriL or in LTriL compared with only five bursts in LVL.
In cat EB8 the fore- and the hindlimbs stepped at the same
frequency and the coupling shifts were within the range of
one-step cycle duration. The ability to maintain a constant
homolateral fore- and hindlimb coupling was permanently
deficient for cat EB7 and transient for cat EB8. Cat EB8 not
only maintained a more constant interlimb coupling but gen-
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on June 18, 2017
FIG. 2. Locomotor deficits during the plateau period. A and C: locomotion of cats EB7 and EB8 in their intact state, using
consecutive stick figure diagrams and angular displacement traces of the left hindlimb together with the raw electromyograms
(EMGs) of the same 5-step cycles. Vertical dotted lines on the angular traces and raw EMGs indicate the moment of foot contact
of the left hindlimb (LH), whereas the duty cycles of that limb are illustrated by the horizontal heavy lines at the bottom. Arrow
heads indicate the foot contact (down) and foot lift (up). B and D: as in A and C, the locomotion of the cat after spinal lesion. To
give an indication of the spatial coordinates, in all the stick figure diagrams, the tibia of cats EB7 and EB8, are 11 and 10 cm,
respectively.
EFFECTS OF NORADRENERGIC AND SEROTONINERGIC DRUGS AFTER PARTIAL SPINAL LESIONS
1517
erally did better long-term postlesion compared with cat EB7.
For example, cat EB8 could follow higher treadmill speeds of
#0.5 m/s, whereas cat EB7 could hardly walk at 0.4 m/s, and
its daily performance was much more variable.
Drug applications
The effects of
NE application, early after the spinal lesion, are illustrated
in Fig. 3 (cat EB7), and the mean changes in the EMG
activity (duration and amplitude) are summarized for both
cats in Table 2.
Early after the lesion (1 wk for cat EB8 and #34 days for cat
EB7), the cats could not support or walk with their hindlimbs,
and they moved around using the forelimbs only. Both cats had
poor lateral stability that resulted in difficulty in righting the
body. However, when light weight support such as holding the
tip of the tail was provided, a few highly disorganized quadrupedal steps could be performed at very low treadmill speeds
(0.1– 0.2 m/s). Such a walking sequence is illustrated for cat
EB7 in Fig. 3B and should be compared with the intact state in
Fig. 3A. It is noticeable from the consecutive stick figures of
the left hindlimb that the cat had difficulty maintaining an
upright position even when the tail was held. The poor walking
NE INJECTION IN THE EARLY PERIOD POSTLESION.
is reflected in irregular angular excursion traces and in highly
disorganized EMG activity. Some episodes of exaggerated
EMG activity resulted from the cat’s attempts to recover from
a stumble. The disorganized stepping pattern of the hindlimbs
is further demonstrated by the foot fall diagram; while the left
hindlimb performed large steps, the right hindlimb performed
short ones. The coupling between the fore- and the hindlimbs
was also highly perturbed. The hindlimbs performed only three
consecutive steps, whereas the forelimbs performed five, indicating that the hindlimbs and the forelimbs walked at different
frequencies of ;0.5 and 1 Hz, respectively.
Twenty minutes after injection of a third bolus of 1.6 mM/
100 ml NE (Fig. 3C) there was a considerable improvement in
the stepping pattern, which became more regular and sustained,
although it still was necessary to maintain part of the weight of
the cat by lightly holding the tail. The improvement in the
walking is evident from the stick figure diagram, which shows
five regular consecutive steps of the left hindlimb at a treadmill
speed of 0.4 m/s. The mean angular displacement measured in
all joints was much improved relative to the predrug condition,
and it was somewhat exaggerated compared with the intact
pattern (see Fig. 3A). For example, after NE application, the
mean angular displacement of the ankle was between 142 6 4
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on June 18, 2017
FIG. 3. Norepinephrine (NE) injection during the early period (cat EB7). Treadmill locomotion of cat EB7 is illustrated using
representative consecutive stick figure diagrams and angular displacement traces of the left hindlimb, the related raw EMGs, and
foot fall diagrams. A: intact condition. B: predrug. C: 20 min after intrathecal injection of 3 consecutive doses of 1.6 mM/100 ml
NE. Dashed vertical lines on the angular traces and raw EMGs correspond to left foot contact. Horizontal lines in the foot fall
diagrams illustrate the period during which each limb is in contact with the treadmill (stance) while the empty spaces correspond
to the swing period. Dashed lines connecting the foot falls illustrate, when possible, the coupling relations between homolateral
fore- and hindlimb. L, left; R, right; H, hindlimb; F, forelimb. Values of the step-cycle duration, normalized EMG amplitude and
bursts duration are summarized in Table 2.
1518
TABLE
E. BRUSTEIN AND S. ROSSIGNOL
2.
Amplitude and duration of EMGs after NE application (early after the spinal lesion)
Intact (0.4 m/s)
Muscle
EB7
LSt
RSt
LSrt
LGM
RGM
LTriL
RTriL
Hindlimb
step cycle
Forelimb
step cycle
LSt
RSt
LSrt
LGM
RGM
LTriL
RTriL
Hindlimb
step cycle
Forelimb
step cycle
EB8
Normalized
Amplitude
100 6 11 (17)
100 6 16 (16)
100 6 10 (17)
100 6 9 (16)
100 6 14 (17)
100 6 9 (16)
100 6 6 (17)
Duration
193 6
215 6
303 6
678 6
581 6
801 6
752 6
19
29
40
64
81
47
77
Normalized
Amplitude
211 6 18† (8)
227 6 36† (7)
137 6 41† (7)
ND
245 6 47† (7)
47 6 21† (6)
111 6 14‡ (8)
1,099 6 57 (15)
100 6 10 (12)
100 6 9 (10)
100 6 12 (12)
100 6 14 (12)
100 6 9 (12)
100 6 10 (12)
100 6 7 (12)
1,098 6
178 6
144 6
383 6
646 6
510 6
741 6
612 6
61 (15)
48
34
50
81
83
77
62
Duration
226 6 111
441 6 228†
466 6 106†
ND
565 6 228
851 6 142
873 6 175‡
Postdrug (0.4 m/s)
Normalized
Amplitude
548 6 22† (12)
302 6 31† (11)
209 6 24† (11)
239 6 35† (11)
113 6 25 (7)
41 6 20† (12)
132 6 11† (12)
2,084 6 816† (4)
89 6 17 (2)
83 6 25 (4)
129 6 15† (4)
ND
75 6 14† (4)
70 6 3† (3)
108 6 12 (4)
1,003 6 189 (7)
76 6 11‡
154 6 51
644 6 267†
ND
1,355 6 171†
1,056 6 307†
1,228 6 216†
Duration
154 6 46†
184 6 52
380 6 58
521 6 83†
401 6 85†
569 6 110†
570 6 110†
934 6 125† (11)
76 6 31† (7)
73 6 18† (7)
133 6 10† (7)
57 6 4† (7)
82 6 25‡ (7)
90 6 13‡ (7)
103 6 7 (7)
770 6
103 6
124 6
369 6
575 6
417 6
616 6
507 6
65† (12)
15†
16
22
77
79‡
55†
54†
1,081 6 170 (12)
1,554 6 635‡ (3)
852 6 63† (7)
1,081 6 193 (12)
1,239 6 223 (4)
854 6 69† (7)
The postdrug time period for cat EB7 was 20 min after three doses of NE (1.6 mM/100 ml) and for cat EB8 was 160 min after one dose of NE (1.6 mM/100
ml). Postlesion time was 9 and 6 days for cats EB7 and EB8, respectively. Number of average step cycles is enclosed in parentheses. The normalized and averaged
amplitude of EMG bursts pre- and postnorepinephrine (NE) injected in the early period, are given for each one of the cats, as percent of the intact values 6 the
coefficient of variation (cv) and are listed for selected muscles in the fore- and the hindlimbs, together with their burst duration in real time (ms 6 SD). ND 5
not detectable during the experiment although the signal was present in other circumstances. EMG, electromyogram; LSt and RSt, left and right semiteninosus;
LSrt, left sartorius; LGM and RGM, left and right gastrocnemius medialis; LTriL and RTriL, left and right lateral head of the triceps brachii. *This speed was
0.1 m/s for cat EB8. The statistical significance of all the mean values was tested against the intact values, using Student’s t-test (†P , 0.01, ‡P , 0.05).
and 83 6 5° (means 6 SE), compared with 134 6 4 to 98 6
3° in the intact situation. This pronounced increase in angular
displacement was accompanied by a marked increase in the
normalized EMG amplitude of the hindlimb muscles (see Table 2), especially of the flexors, which reached beyond values
obtained at the same walking speed in the intact state. The foreand hindlimb coupling pattern also was stabilized as reflected
by more organized foot fall patterns for all four limbs. It should
be noted that there was still a mismatch, although smaller,
between the number of steps performed by the hindlimbs and
the forelimbs (the walking frequencies in the fore- and the
hindlimb were 1.3 and 1.1 Hz, respectively).
NE improved the regularity, maintenance, and walking
speed of cat EB8 as well. For example, predrug, the step cycle
duration of the fore- and the hindlimbs did not differ significantly; nevertheless, they were highly variable (for mean values 6 SD, see Table 2). After NE, the cycle duration in the
fore- and the hindlimbs showed a better match and a much
reduced variability (variance ratio test, P , 0.01). Quantitative
analyses of the EMG activity and the kinematics showed that
NE effects in cat EB8 (see Table 2) were less pronounced
compared with that observed in cat EB7. For example, in cat
EB8, the mean range of angular excursion increased in all the
joints but to a lesser extent compared with cat EB7 (9, 8, and
3° in the hip, knee, ankle, and metatarsophalangeal (MTP)
joints, respectively), and the EMG amplitude of the hindlimb
muscles was changed only slightly (see Table 2). These quantitative differences are probably a result of the different predrug locomotor capacities of each cat, which correlated with
the extent of the spinal lesion (see Fig. 1).
The effects of NE were similar in cats EB7 and EB8 in all
the experiments done during the early period (2 and 3 experiments, respectively) in accordance with the different time
duration of this period in the two cats (for more details see
Table 1).
Application of NE during the plateau period
The injection of NE in cat EB7 was repeated further during
the plateau period (this experiment was not done in cat EB8).
A representative example for such an experiment is given in
Fig. 4 and summarized in Table 3.
Before the drug (Fig. 4A), the cat could follow a treadmill
speed of 0.3 m/s. But its walking pattern (compared with the
intact state, see Figs. 3A and 2A) was irregular as reflected in
the abrupt changes in the joint angular traces and in the
variability of the EMG activity. The coupling phase between
the fore- and the hindlimbs was also variable as illustrated by
the dashed lines connecting the foot falls of the homolateral
limbs. As with application of NE in the early period (Fig. 3),
NE improved locomotion also when given during the plateau
period. In the two situations, the stepping of the hindlimbs
became much more regular (see foot fall diagrams and
smoother angular excursion traces in Fig. 4B) and there was a
general increase in the EMG amplitude (see Table 3). However, there were some differences between NE effects in the
early and the plateau periods. For example, during the plateau
period, the increase in the joint angular excursion was limited
to the knee joint only (from 12 to 34°) and more importantly,
NE stabilized both the coupling pattern and the phase drifts
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on June 18, 2017
Cat
Predrug (0.2 m/s)*
EFFECTS OF NORADRENERGIC AND SEROTONINERGIC DRUGS AFTER PARTIAL SPINAL LESIONS
1519
between the homolateral fore- and the hindlimbs (see foot falls
diagrams in Fig. 4B) to a pattern which resembles the intact
one (Fig. 3A).
Clonidine
In light of the improvement in the walking after injection of
NE, we further tested the a2-noradrenergic agonist, clonidine
because it is implicated in the initiation and the modulation of
locomotion in the completely transected spinal cats (for review
see Rossignol 1996). The effects of intrathecal injection of
clonidine in the early period were quite different from those
observed in the complete spinal cat, and they were also different from the effects of NE in the partial spinal cat at that period
(see Fig. 3). Not only clonidine did not improve the locomotor
capability of the cats, but it appeared to impair it. However,
because the predrug pattern was highly deficient (see Fig. 3B)
it was difficult to assess the changes quantitatively.
In the plateau period, the adverse effects of clonidine were
very obvious in both cats. At that stage, even one bolus of
clonidine (0.8 –1.9 mM/100 ml) was sufficient to cause a major
reduction in the weight support of the hindlimbs, to increase
swaying of the hindquarters, stumbling, and falling, which
limited the cats to a few steps at a time at very low treadmill
speeds (0.2 m/s). The effects of clonidine on the locomotion of
cat EB7 are illustrated in Fig. 5B.
Forty-five minutes after injection of clonidine (1.9 mM/100
ml) the cat could barely support its hindquarters and could
hardly step. As illustrated in Fig. 5B, the stick figure diagram,
the joint angular excursion traces and the EMG activity, are
extremely disorganized (see Fig. 6A for the average changes in
EMGs amplitude). These detrimental effects were long-lasting
and the return to predrug walking capacities could not be
observed for the following five recording hours.
The effect of clonidine could be reversed mostly by an
injection of yohimbine (a2-noradrenergic antagonist). As illustrated in Fig. 5C, 60 min postinjection of yohimbine (20.5
mM/100 ml), given 45 min postinjection of clonidine, thus well
within the period of maximal action of clonidine (Fig. 5B), the
weight support of the hindquarters improved and so did the
regularity of the stepping, and the cat’s ability to walk at higher
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on June 18, 2017
FIG. 4. NE injection during the plateau period (cat EB7). Representative walking sections taken at the same treadmill speed (0.3
m/s) to compare the locomotion of cat EB7 in A, predrug, and in B, 20 min after intrathecal injection of 3.2 mM/100 ml NE (for
details see Fig. 3). Related averaged bursts amplitude and duration are given in Table 3.
1520
TABLE
E. BRUSTEIN AND S. ROSSIGNOL
3.
Amplitude and duration of EMGs after NE application (plateau period)
Predrug
Cat
Muscle
EB7
LSt
LSrt
RSrt
LGM
RGM
LTriL
RTriL
Hindlimb step cycle
Forelimb step cycle
Normalized
Amplitude
100 6 30 (18)
100 6 13 (18)
100 6 15 (16)
100 6 6 (17)
100 6 43 (17)
100 6 9 (18)
ND
Postdrug
Duration
175 6 78
327 6 95
250 6 53
685 6 135
538 6 87
659 6 119
Normalized
Amplitude
154 6 23* (18)
105 6 16 (18)
187 6 18* (18)
148 6 13* (18)
310 6 25* (18)
114 6 11* (17)
1,065 6 152 (18)
931 6 108 (18)
Duration
130 6
300 6
240 6
614 6
470 6
600 6
48†
51
34
61
62†
75
864 6 74* (18)
871 6 113 (17)
The normalized and averaged EMG amplitude pre- and post-NE injection with their mean duration, during the plateau period (same format as Table 2, except
that the postdrug values are tested statistically against the predrug values). Postdrug period was 20 min after NE injection (3.2 mM/100 ml) 104 days post lesion
(0.3 m/s). *P , 0.01. †P , 0.05.
Methoxamine
Because there was a major locomotor improvement after
injection of NE, but none with the a2-noradrenergic agonist,
clonidine, we further tested the effects of an a1-noradrenergic
agonist, methoxamine on the locomotion (applied only during
the plateau period). The results of such experiments, are illustrated in Figs. 7 and 8, and the EMG values are summarized in
Fig. 9. These results with methoxamine were reproduced in
both cats in all the experiments (for details, see Table 1).
Methoxamine considerably improved the locomotion of cat
EB7. An hour and 30 min after application of 4 mM/100 ml
methoxamine, the steps were more robust and regular, with
forceful foot placement and were performed with a better
lateral stability. The regularity of the steps is noted clearly
when comparing the consecutive stick figures diagram before
(Fig. 7A) and after methoxamine (Fig. 7B). The related angular
excursion traces show that the movements of the joints are
smoother and reproducible from cycle to cycle because all the
abrupt changes characteristic of the predrug walking disappeared. The average range of movement at the hip, ankle, and
MTP joints decreased after the application of methoxamine
(from 20 down to 14°, 33 to 22°, and from 57 to 44°, respectively) whereas a slighter decrease was observed in the range of
the knee movement (from 27 to 23°). The activity of the
hindlimb muscles markedly changed. There was a general
increase in the normalized amplitude compared with the predrug situation (see especially RGM). In addition, the duration
of LSt increased significantly (see average EMG values in
histograms of Fig. 9).
FIG. 5. Injection of clonidine followed by yohimbine (cat EB7). Treadmill stepping sequences taken 210 days postlesion. A:
predrug; B and C: postdrugs. Consecutive stick figures of the left hindlimb, the angular displacement and the raw EMGs in B
illustrate that, 45 min after intrathecal injection of 1.9 mM/100 ml clonidine, the cat sagged and could not maintain the weight of
the hindquarters nor walk. However, as illustrated in C, 60 min after intrathecal injection of 20.5 mM/100 ml yohimbine, given 45
min after clonidine, the walking of the cat improved considerably as noted from regular stick figure diagrams, joint angular
displacement traces and raw EMG taken at treadmill speed of 0.3 m/s.
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on June 18, 2017
speeds (#0.3 m/s). Yohimbine also improved the regularity of
the EMG pattern, which returned to predrug level (for quantitative effects on the EMG activity in both cats, see Fig. 6).
EFFECTS OF NORADRENERGIC AND SEROTONINERGIC DRUGS AFTER PARTIAL SPINAL LESIONS
1521
After the administration of methoxamine, the ability of the
cat to follow higher treadmill speeds did not improve. Nevertheless there was a major improvement in the ability of the cat
to maintain locomotion at the same speed range for longer
periods of time. Predrug, the cat could make about 6 consecutive steps, whereas after application of methoxamine, it easily
did 20 steps.
The coupling phase between the homolateral fore-and hindlimb also changed after the drug application, as illustrated in
Fig. 8. The graphs of the phase of burst onset of LTriL and
LVL show that, predrug (Fig. 8B), the homolateral coupling
phase varied between 0.4 – 0.8, whereas post methoxamine
(Fig. 8C), it was within the intact range (Fig. 8A) and fluctuated
around a value of 0.25. Further, the coupling pattern was
changed from a diagonal type predrug (Fig. 8B) to almost an
intact pattern with the difference that the forelimbs were leading. The general improvement in the locomotion after methoxamine was long-lasting and could be observed for 4 –5 h.
Methoxamine improved the walking of cat EB8 as well (for
the average EMG values see Fig. 9). As with cat EB7, the
regularity of the hindlimbs stepping improved. This is indicated by a decrease in the variability of the hindlimb step cycle
duration (variance ratio test P , 0.01, see mean values in Fig.
9D). There was also a stabilization of the coupling between the
homolateral fore- and hindlimbs, around a phase value of 0.2,
which resembles the intact situation. Methoxamine also caused
a decrease in the average range of angular displacement in the
hip, knee, and ankle joints (from 39 down to 33°, 38 to 26°, 30
to 19°, respectively), much like what was observed in cat EB7.
These changes, in contrast to the observed in cat EB7, were
accompanied by a decrease in EMG amplitude (see Fig. 9).
Serotoninergic drugs
Several serotoninergic drugs were tested: the neurotransmitter itself, 5HT; the 5HT1,2,3 agonist, quipazine; the 5HT precursor, 5-hydroxytryptophan (5HTP); and a 5HT1A agonist,
8-hydroxy-dipropylaminotetralin (DPAT). All these drugs, except the agonist DPAT, improved locomotion in a similar
manner. They differed mainly in their time of action; 5HT and
5HTP did not last long (,2 h), whereas the effects of quipazine
could be observed for #5 h after the application. DPAT was
detrimental to locomotion of EB7 by inducing, as soon as 20
min postapplication, a severe foot drag of both hindlimbs
causing the cat to stumble very often. This effect faded away
after 1 h.
A representative example for the effects of quipazine on the
locomotion of cat EB7 is given in Figs. 10 and 11, and the
quantitative values of the EMG activity are summarized in the
histograms of Fig. 12. Predrug, the cat walked irregularly, as
illustrated in the stick figure diagram (Fig. 10A) showing five
consecutive steps, each of a different duration. In addition, the
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on June 18, 2017
FIG. 6. Histograms of EMG amplitude and burst duration after clonidine and clonidine followed by yohimbine (cat EB7 and
EB8). Pairs of histograms illustrate, on the left, the average and normalized EMG amplitude expressed as percent of the predrug
values 6 coefficient of variation (cv). Right: related averaged burst duration 6 SD. Time postdrug injection and their doses: A and
B: cat EB7, 45 min after 1.9 mM/100 ml clonidine and 60 min after 20.5 mM/100 ml yohimbine at day 210 postlesion. C and D:
cat EB8, 30 min after 1.9 mM/100 ml clonidine and 60 min after 20.5 mM/100 ml yohimbine, at day 98 postlesion. Pre, predrug;
Post, postdrug; HCD, hindlimb cycle duration; FCD, forelimb cycle duration; C 1 Y, injection of yohimbine after injection of
clonidine; ND, not detectable. Statistical significance of the changes in EMG amplitude and duration was tested using Student’s
t-test (*P , 0.05, **P , 0.01).
1522
E. BRUSTEIN AND S. ROSSIGNOL
angular movement of the joints was variable and included
abrupt, sudden changes. The raw EMGs also reflected this
disorganized walking. The EMG activity of the different muscles not only varied in amplitude and duration but also in their
time course (see especially right and left GM).
After injection of quipazine (1.2 mM/100 ml), the cat’s
walking improved considerably. It had better lateral stability
and it could take long regular steps at a constant speed of 0.3
m/s (see Fig. 12), which were, still, shorter than the mean intact
step cycle duration at 0.3 m/s (1,372 6 62 ms). After the drug,
the angular excursion traces were more regular and stable. In
addition, there was a major increase in the average range of
angular excursion of the hip, knee, and ankle: from 17 to 27°,
21 to 29°, and from 19 to 28°, respectively. The values of
angular excursion, after quipazine, resembled more the intact
values (26° in the hip, 32° in the knee, and 36° in the ankle).
No change was observed in the average range of angular
excursion of the MTP joint, which stayed lower relative to the
intact. The EMG activity also reflected the increase in regularity and prolongation of the steps. There was a general
tendency for prolongation of the burst duration relative to the
predrug situation (see Fig. 12B), especially in the extensors.
The latter was even significantly longer then the intact burst
duration. Quipazine also stabilized the interlimb coupling, as
illustrated in the average foot fall diagrams in Fig. 11. Predrug
the coupling pattern was variable, as expressed by a continual
shift between three main types of coupling patterns, homolateral in-phase coupling, diagonal coupling (Fig. 11A), and almost a normal coupling pattern (see average foot fall diagram
of the intact in Fig. 8A). After quipazine, the cat was walking
consistently with an homolateral, in-phase, coupling pattern,
which is illustrated in the foot fall diagram in Fig. 11B,
averaging 13 consecutive steps. This more robust coupling
pattern probably resulted from an improved hindlimb stepping
regularity (variance ratio test P , 0.01) and not from a better
coupling between the fore- and the hindlimbs because even
postdrug, the drift in the coupling phase persisted (see the
related phase plot).
Combination of methoxamine and quipazine
As detailed in the preceding sections, both methoxamine and
quipazine improved locomotion but in a different way. The
first drug increased the amplitude of the EMGs, decreased the
cycle duration and the joint angular excursion while the later
prolonged the step cycle duration. Therefore both drugs were
given at the same time in one bolus. The effects on stability of
the walking is illustrated in Fig. 13, whereas the quantitative
values are given in Fig. 14.
The cat walking (regularity, maintenance, and adaptation to
speeds) is illustrated in Fig. 13 by plotting the step cycle
duration of a whole consecutive stepping sequence performed
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on June 18, 2017
FIG. 7. Injection of methoxamine during the plateau period (cat EB7). Representative walking sections taken at the same treadmill
speed (0.3 m/s) to compare the locomotion of cat EB7 in A, predrug, and in B, 80 min after intrathecal injection of 4 mM/100 ml
methoxamine (for details see Fig. 1). Mean values of bursts amplitude and duration are given in the histograms of Fig. 9.
EFFECTS OF NORADRENERGIC AND SEROTONINERGIC DRUGS AFTER PARTIAL SPINAL LESIONS
1523
during one recording session. Before the drug (Fig. 13B), the
cat had major difficulty maintaining a treadmill speed of 0.35
m/s for .10 step cycles, and after several steps in which it
touched the back of the treadmill, the speed of the treadmill
was decreased to 0.3 m/s. Even then the cat walked irregularly
as reflected by a step to step fluctuation in the cycle duration.
Sixty-five minutes postapplication of one bolus of the combined solution, which contained only half the doses (2 mM
methoxamine 1 1.2 mM quipazine/100 ml) used when each
one of the drugs was tested separately, an elegant and sustained
walking was observed, with much improved lateral stability.
This is indicated by a much reduced variability of the step
cycle duration, which resembled generally the intact situation
(Fig. 13A). Notice, however, that in the intact situation, walking at 0.3 m/s was somewhat too slow for this cat, and this was
expressed in higher step cycle variability. In addition to improve regularity there was also an improvement in the speed
range the cat could follow and maintain (#0.4 m/s). Generally,
the stepping was characterized by somewhat shorter steps (see
Fig. 14B) and by a corresponding slight reduction in the
average range of the angular displacement (not illustrated). The
EMG duration and amplitude showed various changes as summarized in Fig. 14, A and B.
The application of the combined solution also improved cat
EB8’s walking. In this cat we have observed an increase in the
step cycle duration (see Fig. 14D). Changes also were found in
the average range of angular excursion of the hip, ankle, and
MTP, which decreased from 29 to 23°, 29 to 21°, and 48 to
43°, respectively. Significant increases were found in the amplitudes of all the flexors (see Fig. 14, C and D). The extensor
amplitudes did not change, but their duration increased significantly compared with the predrug situation.
DISCUSSION
In this work, we have demonstrated that different noradrenergic and serotoninergic drugs can improve or deteriorate
walking of cats with partial but large ventral and ventrolateral
lesions of the low thoracic spinal cord. First the extent of the
spinal lesion with special emphasis on the damage to descending noradrenergic and serotoninergic tracts will be discussed.
Then different possible mechanisms by which the drugs may
exert their effects will be suggested as well as possible clinical
applications.
Extent of the spinal lesions in relation to descending
noradrenergic and serotoninergic pathways
The extent of the damage to noradrenergic and serotoninergic pathways after the ventral and ventrolateral spinal lesions
only could be evaluated indirectly by comparing the results of
the evaluation of the extent of the spinal lesion (for details, see
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on June 18, 2017
FIG. 8. Interlimb coupling after methoxamine injection (cat EB7). A: intact situation. B: predrug. C: 80 min after intrathecal
injection of 4 mM/100 ml methoxamine. Data used to illustrate the coupling relations are taken from the pre- and postdrug walking
sections illustrated in Fig. 7. Graphs illustrate the homolateral fore- and hindlimb coupling using the phase values of burst of an
extensor muscle in the fore- and the hindlimb, TriL, and VL (relative to LSt), respectively, for several consecutive step cycles. The
normalized foot falls at the bottom of each graph demonstrate the average coupling pattern the cat used in each walking sequence.
Filled bars represent the stance phase of each one of the limbs. Vertical lines divide the average step cycle according to the number
of supporting limbs. For each division, the related figurines demonstrate which limb was in contact with (filled circles) or lifted of
the ground.
1524
E. BRUSTEIN AND S. ROSSIGNOL
the preceding paper, Brustein and Rossignol 1998) to what is
known from the literature about the noradrenergic and serotoninergic pathways in the spinal cord.
Previous studies using histofluorescence and retrograde HRP
labeling showed that noradrenergic pathways originating in
nucleus locus coeruleus, the main source of spinal noradrenergic fibers, project primarily through the ventrolateral funiculus. In addition, nucleus subcoeruleus, nucleus Kolliker-Fuse,
and cell bodies in A5 send fibers through both the ventrolateral
and the dorsolateral funiculi. The noradrenergic fibers innervates the lumbar spinal cord bilaterally, although the ipsilateral
contribution dominates (Kuypers 1981; Kuypers and Maisky
1977; Marshall 1983; Stevens et al. 1985). The serotoninergic
axons originating in the raphe nuclei, such as raphe pallidus
and obscuras, descend mainly in the ventral and ventrolateral
funiculi, whereas axons from raphe magnus descends primarily
through the DLF (Kuypers 1981; Martin et al. 1978).
According to these findings, most of the noradrenergic and
the serotoninergic axons descending in the ventral and ventrolateral quadrants of cat EB7 probably were damaged. However,
the noradrenergic and the serotoninergic axons descending in
the left DLF mostly were preserved. In cat EB8, the existence
of serotoninergic axons in left ventral quadrant cannot be
excluded. Nevertheless, an extensive damage probably was
done to serotoninergic and noradrenergic fibers descending in
the lateral funiculi. The difference in the extent of the spinal
lesions (less severe for cat EB8), which correlated with faster
recovery and better locomotor performance (see Brustein and
Rossignol 1998), may explain why generally larger doses of
the tested drugs were needed to see effects in cat EB8 long-
term postlesion (see Table 1). Yet, and most importantly, the
applied drugs showed overall similar effects in both cats.
Drugs such as clonidine had a detrimental effect in both cats,
and drugs that caused a pronounced improvement in the walking of cat EB7 may have caused more subtle effects on the
walking of cat EB8, but they were never detrimental.
Noradrenergic drugs injection in the partial spinal cat
Injection of the neurotransmitter
itself, NE, in the early period, highly improved speed performance and stepping regularity of both cats. It also improved
the hindlimb joint angular excursion and the coupling between
the homolateral fore- and hindlimb. These effects were more
pronounced in cat EB7 than in cat EB8 most likely due to
differences in their locomotor capacities, which related to the
extent of the spinal lesions. In the early period, NE improved
the stepping pattern considerably but not to the point that cats
could walk on their own with full-weight support and lateral
stability of the hindquarters. In the plateau period, however, the
improvement in the walking regularity was accompanied by a
pronounced increase in the weight support and lateral stability.
The latter effects resembled the effects of a1-agonist, methoxamine, and may be attributed to an increase in the extensor
amplitude. Methoxamine, in contrast to NE, caused a general
decrease in the joint angular excursion, which may explain
why no improvement was observed in speed performance.
Nevertheless, better stability, more regular hindlimb stepping,
and interlimb coupling may account for more sustained walking at the same range of speeds. It is interesting to note that,
SUMMARY OF THE EFFECTS.
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on June 18, 2017
FIG. 9. Histograms of EMG amplitude and burst duration after methoxamine injection (cats EB7 and EB8). Same format as Fig.
6. Time postdrug injection and their doses: A and B: cat EB7, 80 min after 4 mM/100 ml methoxamine 196 days postlesion. C and
D: cat EB8, 50 min after 6 mM/100 ml methoxamine at day 104 postlesion (*P , 0.05, **P , 0.01).
EFFECTS OF NORADRENERGIC AND SEROTONINERGIC DRUGS AFTER PARTIAL SPINAL LESIONS
1525
after methoxamine, cat EB8 showed more regular walking and
a decrease in joint angular excursion, as did cat EB7, but
without major effects on the amplitude of the recorded muscles.
Contrary to the locomotor improvement seen after NE and
the a1-agonist, methoxamine, the a2-agonist, clonidine, caused
deterioration of the walking in both cats. There was a major
decrease in weight support and in lateral stability of the hindquarters, which resulted in very deficient and inconsistent
stepping at low treadmill speeds.
PROPOSED MECHANISMS FOR THE ACTIONS OF THE DRUGS.
The
improvement in locomotion of the partial spinal cat after NE
and methoxamine injections can be attributed to global changes
in spinal neurons excitability (Grillner 1981). According to this
suggestion, the slight hyperpolarization caused by NE, a result
of a decrease in membrane conductance, will cause a general
potentiation of 25–50% of other non-NE synaptic input. The
possible implication of noradrenergic drugs in mechanisms of
gain amplification was suggested as well by Hounsgaard et al.
(1988) and by Conway et al. (1988). L-3-4-Dihydroxyphenylalanine (L-DOPA) and clonidine induce plateau potentials in
spinal motorneurons of acute complete spinal cats (Conway et
al. 1988; Schomburg and Steffens 1996), so does methoxamine
in the decerebrate cat preparation (Lee and Heckman 1996,
1997). Plateau potentials were proposed as a mechanism that
reduces the need for sustained synaptic drive from the rhythm
generator during locomotion. Therefore by maintaining excitability at a constant level, only short-lasting excitation or
terminating inhibition is needed to start or stop the activity in
the motorneurons (Conway et al. 1988; Hounsgaard et al.
1988). This may account for the major improvement in the
walking regularity after methoxamine and NE in both cats even
without major changes in EMG amplitudes as observed in cat
EB8. Hence NE and methoxamine probably provided a constant background excitability on which the signals from the
rhythm generator could be expressed reliably and consistently.
In contrast to NE and methoxamine, application of clonidine
in the partially spinal cat had a detrimental effect on the
walking. These results are also in contrast to those observed in
early chronic spinal cats in which clonidine initiates hindlimb
locomotion (Barbeau and Rossignol 1991; Barbeau et al. 1993;
Chau et al. 1998b). In the late complete spinal cat (Barbeau et
al. 1993; Chau et al. 1998b), small doses of clonidine given
intrathecally (1 mg/100 ml), modulated the stabilized spinal
hindlimb walking, increase of step length, joint angular displacement, and the amplitude and duration of flexors. However, larger dose of clonidine (10 mg/100 ml it), decreased the
amplitude of the extensors, reduced the weight support of the
hindlimbs, and induced a pronounced foot drag, much like in
the partial spinal cat. For the partially spinal cat, these effects
were much more severe because the cats hardly could continue
walking quadrupedally.
The detrimental effect of clonidine on the walking of partial
spinal cats, its beneficial effects in the early spinal cats, and its
dose-dependent effects in late complete spinal cats could be
explained by a difference in the a2-receptor population remaining in each condition. After a complete spinal transection, there
is a removal of the central presynaptic a2 receptors and the
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on June 18, 2017
FIG. 10. Injection of quipazine (cat EB7). Consecutive stick figure diagrams, joint angular traces of the left hindlimb and the
corresponding EMG activity of selected muscles in the fore- and the hindlimbs (for details see Fig. 1) to illustrate the walking
capacities of cat EB7. A: predrug; B: 75 min after 1.2 mM/100 ml quipazine. Mean amplitude and duration of EMG bursts are
summarized in Fig. 12 in an histogram form.
1526
E. BRUSTEIN AND S. ROSSIGNOL
effect of the drugs is mainly due to activation of the remaining
a2 postsynaptic receptors (Langer 1977; Marshall 1983). In the
partially spinal cats, some presynaptic a2 receptors must be
spared on remaining descending adrenergic terminals, mainly
in the dorsal horn where they are mostly concentrated (Giroux
et al. 1995) and also on primary sensory afferents (Howe et al.
1987). The latter may be even larger in number after the partial
lesion, a result of sprouting of primary afferents (Helgren and
Goldberger 1993). Thus in the partial spinal cat, in contrast to
the complete spinal cat, application of clonidine also could
largely affect presynaptic receptors to cause a decrease in NE
release.
There are some evidences that indeed, L-DOPA, the noradrenergic precursor, depressed transmission of activity from
primary afferents but not of group I (Anden et al. 1966).
Further, the local application of NE and NE agonists also was
found to depress transmission in interneuronal pathways from
group II afferents (Bras et al. 1989, 1990; Jordan et al. 1977),
an effect that was observed in populations of neurons at L4 –5,
where the tip of the intrathecal cannula is located in both cats.
It is interesting to note that the depressive effect was primarily
exerted by a2-agonists such as tizanidine. The depressive effects of descending noradrenergic system on interneuronal
pathways involved in mediating the reflex action of group II
afferents on motorneurons was demonstrated further by stimulation of the locus coeruleus and subcoeruleus of decerebrated
cats (Jankowska et al. 1993). Further, it was reported that
L-DOPA reduced spasticity in paraplegic patients (Eriksson et
al. 1996).
The aforementioned effects of noradrenergic drugs may not
be beneficial for the partial spinal cat, as a reduction in the
transmission from the periphery may interfere with the potential compensatory mechanisms from sensory feedback that the
cats may develop after the spinal lesion to maintain weight
support and locomotion (Brustein and Rossignol 1998; Grillner
1972; Guertin et al. 1995).
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on June 18, 2017
FIG. 11. Interlimb coupling after quipazine injection (cat EB7). Data used to illustrate the coupling relations are taken from the
walking sections of Fig. 10 and correspond to predrug (A) and 75 min after intrathecal injection (B) of 1.2 mM/100 ml quipazine
(for details see Fig. 8). Notice, however, that in A, 3 average foot fall diagrams are illustrated to show the variability in the coupling
pattern used by the cat in the predrug situation. L, left; R, right; H, hindlimb; F, forelimb.
EFFECTS OF NORADRENERGIC AND SEROTONINERGIC DRUGS AFTER PARTIAL SPINAL LESIONS
1527
The important effect of NE and methoxamine on the coupling between the fore- and hindlimbs may be, as discussed
earlier, due to an improved background excitability level, on
which the signals from the rhythm generator can be expressed
more consistently. Stabilization of the hindlimb walking pattern will result in a generally more organized quadrupedal
coupling. It is also possible that the observed improvement in
the hindlimb locomotion results in a more consistent information flow between the two girdles through the remaining ascending and descending spinal pathways to better integrate
their activity. Such a modulatory action (facilitatory and inhibitory) of NE on ascending information from muscle and skin
afferents was suggested by Jankowska et al. (1997).
Serotoninergic drugs injection in the partial spinal cat
5HT, the neurotransmitter itself,
its precursor, 5HTP, and quipazine, a 5HT1,2,3 agonist, improved the lateral stability and the weight support of the
hindlimbs, which resulted in a more regular quadrupedal walking with better interlimb coupling pattern, especially in the
most lesioned cat, EB7. In addition, the serotoninergic drugs
prolong the step cycle duration in both cats. In cat EB8, this
prolongation was accompanied by an increase in flexors and by
a slight increase in extensor burst duration, whereas in cat EB7,
a significant increase in extensor burst duration was observed
with a slight decrease in the normalized EMG amplitude.
These effects were different from those observed after application of NE or methoxamine, which decreased the step cycle
SUMMARY OF THE EFFECTS.
duration, joint angular excursion, and increased the amplitude
of the EMGs.
The 5HT1A agonist, DPAT, had a detrimental effect on the
locomotion of cat EB7, which was manifested as a severe foot
drag in both hindlimbs, causing the cat to stumble often.
However, compared with clonidine, this effect was short-lasting and was not accompanied by an apparent decrease of the
weight support of the hindlimbs or an increased wobbliness.
In
contrast to NE and clonidine, the 5HT precursor, 5HTP, was
not found to evoke locomotor rhythm in either the low spinal
and decerebrate cat (Grillner and Shik 1973) or in the complete
chronic spinal cat during the first week after spinalization,
whereas clonidine or L-DOPA caused a dramatic change in
kinematics and EMG pattern (Barbeau and Rossignol 1990,
1991; Barbeau et al. 1993). However, 5HTP markedly increased the tonic activity in all muscles in both preparations
(Barbeau and Rossignol 1991; Grillner and Shik 1973). An
increase in extensor and flexor muscles activity (amplitude and
duration) also was observed in the chronic late spinal cat after
intraperitoneal injection of different serotoninergic drugs (Barbeau and Rossignol 1990, 1991).
The involvement of serotonin in the modulation of the
membrane properties of motorneurons in the decerebrate cat
was demonstrated by Hounsgaard et al. (1988). It was shown
that the bistable properties, mainly of extensor motorneurons,
which disappeared after acute spinalization, could be restored
after intravenous application of large doses of 5HTP. The
PROPOSED MECHANISMS FOR THE ACTIONS OF THE DRUGS.
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on June 18, 2017
FIG. 12. Histograms of EMG amplitude and burst duration after injection of serotoninergic drugs (cats EB7 and EB8). Same
format as Fig. 6. Time postdrug injection and their doses: A and B: cat EB7, 75 min after 1.2 mM/100 ml quipazine, 227 days
postlesion. C and D: cat EB8, 150 min after 2.3 mM/100 ml serotonin precursor (5HTP) at day 168 postlesion (*P , 0.05, **P ,
0.01).
1528
E. BRUSTEIN AND S. ROSSIGNOL
effect of 5HTP on the bistable properties of the motorneurons
have many similarities to the effect of L-DOPA (Conway et al.
1988). However, they differ in their targets and strength of
action. 5HTP strongly affected extensor motorneurons,
whereas L-DOPA affected extensors and flexors but to a lesser
extent. This may explain why after 5HT we observed a tendency to increase step cycle duration and bursts duration,
especially in extensors (see cat EB7), whereas after application
of NE, early after the lesion, a general increase was noted in the
amplitude of all muscles, especially in flexors. During the
plateau period, however, the effects of NE and methoxamine
on extensors were more pronounced compared their effects
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on June 18, 2017
FIG. 13. Duration of consecutive step cycles after injection of methoxamine and quipazine. Cycle duration (ms) of consecutive steps taken during 1
experimental session at different treadmill speeds (indicated by the arrows) are
illustrated for cat EB7. A: intact situation. B: predrugs. C: 75 min after
combined injection of methoxamine and quipazine (2 mM and 1.2 mM/100 ml,
respectively).
during the early period, probably a result of different excitability levels. In the early period, extensor activity was almost
absent, a result of elimination of strong excitatory input from
descending pathways. At that stage, a very strong excitatory
input is needed to increase their weak activity. In contrast,
long-term postlesion, when some adaptive processes already
have taken place (see Brustein and Rossignol 1998 for discussion), the baseline activity of these muscles is much better and
less input is needed to increase their action. This possible
explanation is supported by observations that the level of the
bias current applied to extensor motorneurons in the decerebrate cat had an important effect on expression of the bistable
properties (Hounsgaard et al. 1988).
The involvement of serotoninergic drugs in restoring tonic
background activity in extensors was further demonstrated by
Miller et al. (1996). 5HT2 agonist, 2,5-dimethoxy-4-iodoamphetamine (DOI), restored the tonic background activity in GM
and soleus and facilitated the GM stretch reflex, both abolished
after acute spinalization of the decerebrate cats. Such a tonic
excitation was suggested by Mori (1987) to be the main action
of serotoninergic system.
5HT, like NE, also was found to have a specific modulatory
effect on transmission from peripheral afferents (Bras et al.
1989, 1990; Jankowska et al. 1993, 1997). 5HT was found to
depress transmission from group II afferents but not of group
I afferents. However, its effectiveness was different than seen
after NE. NE had more pronounced effects in depressing
potentials in the ventral horn, and its action appeared sooner
and was faster relative to 5HT (Bras et al. 1989). It further was
suggested that this depression may involve different membrane
receptors at different locations. a2-agonists were found to be
effective primarily in the intermediate zone and in the ventral
horn, whereas 5HT1A was effective in the dorsal horn (Bras et
al. 1990). Such influences were observed as well after stimulation of the raphe nucleus (Jankowska et al. 1993). 5HT also
was found to affect the ascending transmission from group II
afferents (Jankowska et al. 1997). These modulatory effects
may explain why in cat EB8, which had better weight support,
5HT drugs improved the regularity of the locomotion without
a major effect on EMG activity.
Both quipazine and methoxamine given separately had beneficial effects on locomotion. Their combination was most
efficient in improving the locomotion of the cats, and only half
the dose of each drug used individually was sufficient to give
a pronounced effect. The combination of both drugs induced
the most regular and maintained locomotion, with elegant and
smooth movements, probably because they complete each other’s actions.
Taken together it seems that the serotoninergic and the
noradrenergic systems can modulate the pattern of locomotion
by changing the neuronal excitability. In the absence of, or
after a decrease in, these descending inputs, signals for generation of the walking rhythm cannot be expressed consistently,
resulting in irregular walking. A lesion to these descending
pathways effect as well the coupling between the fore- and the
hindlimbs. It also seems that serotoninergic and noradrenergic
neurotransmitters contribute to continual stabilization of the
fore- and hindlimb coupling by modulating the transmission of
ascending information from the muscle and skin afferents.
Our results have important clinical implications, showing
that the effects of some drugs such as clonidine depend on
EFFECTS OF NORADRENERGIC AND SEROTONINERGIC DRUGS AFTER PARTIAL SPINAL LESIONS
1529
whether or not the spinal lesion is complete. In the complete
spinal cat, clonidine may exert beneficial effects; however,
in the partial spinal cats, its effects may interfere with
compensatory mechanisms developed after the lesion. This
may indicate that different classes of drugs may be used in
patients with different types of spinal cord injuries. Indeed,
results of intrathecal injection of clonidine in patients
showed an important variability of effects (Barbeau et al.
1998; Dietz et al. 1995). In addition, a most important
finding is that these drugs effects can be integrated into the
residual voluntary locomotor control to improve locomotion
and its postural aspects. It is then probable that such changes
in neuronal excitability could be used beneficially by patients who still have residual capabilities to induce the
rhythm of walking but are unable to sustain it.
We gratefully acknowledge the contribution of J. Provencher and F. Lebel
for assistance during surgeries, experiments, analyses, and preparation of the
illustrations; P. Drapeau and G. Messier for programming; J. Lavoie, F. Cantin,
and N. De Sylva for histological assistance; C. Gauthier and D. Cyr for
illustrations and photographs; and C. Gagner for electronic support.
This work was supported by the Canadian Neuroscience Network and a
group grant from the Medical Research Council of Canada. E. Brustein was
supported by fellowships from the Canadian Neuroscience Network and the
Groupe de Recherche sur le Système nerveux central (Fonds pour la Formation
de Chercheurs et l’Aide à la Recherche).
Address for reprint requests: S. Rossignol, Centre de recherche en Sciences
Neurologiques, Université de Montréal, P.O. Box 6128, Station Centre-Ville,
Montreal, Quebec H3C 3J7, Canada.
Received 4 February 1998; accepted in final form 25 November 1998.
REFERENCES
ANDEN, N. E., JUKES, M. G., LUNDBERG, A., AND VYKLICKY, L. The effect of
DOPA on the spinal cord. I. Influence on transmission from primary afferents. Acta Physiol. Scand. 67: 373–386, 1966.
BARBEAU, H., CHAU, C., AND ROSSIGNOL, S. Noradrenergic agonists and
locomotor training affect locomotor recovery after cord transection in adult
cats. Brain Res. Bull. 30: 387–393, 1993.
BARBEAU, H., NORMAN, K., FUNG, J., VISINTIN, M., AND LADOUCEUR, M. Does
neurorehabilitation play a role in the recovery of walking in neurological
population? In: Neuronal Mechanisms for Generating Locomotor Activity,
edited by O. Kiehn, H. Hultborn, N. Kudo, L. M. Jordan, and R. M.
Harris-Warrick. New York: Ann. NY Acad. Sci., 1998, p. 377–392.
BARBEAU, H. AND ROSSIGNOL, S. The effects of serotonergic drugs on the
locomotor pattern and on cutaneous reflexes of the adult chronic spinal cat.
Brain Res. 514: 55– 67, 1990.
BARBEAU, H. AND ROSSIGNOL, S. Initiation and modulation of the locomotor
pattern in the adult chronic spinal cat by noradrenergic, serotonergic and
dopaminergic drugs. Brain Res. 546: 250 –260, 1991.
BARBEAU, H. AND ROSSIGNOL, S. Enhancement of locomotor recovery following spinal cord injury. Curr. Opin. Neurol. 7: 517–524, 1994.
BEM, T., GORSKA, T., MAJCZYNSKI, H., AND ZMYSLOWSKI, W. Different patterns
of fore-hindlimb coordination during overground locomotion in cats with
ventral and lateral spinal lesions. Exp. Brain Res. 104: 70 – 80, 1995.
BRAS, H., CAVALLARI, P., JANKOWSKA, E., AND MCCREA, D. A. Comparison of
effects of monoamines on trasmission in spinal pathways from group I and
II muscle afferents in the cat. Exp. Brain Res. 76: 27–37, 1989.
BRAS, H., JANKOWSKA, E., NOGA, B., AND SKOOG, B. Comparison of effects of
various types of NA and 5HT agonists on transmission from group II muscle
afferents in the cat. Eur. J. Neurosci. 211: 1029 –1039, 1990.
BRODAL, A. Neurological Anatomy. London: Oxford Univ. Press, 1969.
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on June 18, 2017
FIG. 14. Histograms of amplitude and duration of bursts (cats EB7 and EB8) after a combined injection of methoxamine and
quipazine. Same format as Fig. 6. Time postdrug injection and their doses: A and B: cat EB7, 65 min after (2 mM 1 1.2 mM)/100
ml methoxamine and quipazine, respectively, at 245 days postlesion. C and D: cat EB8, 75 min after (4 mM 1 2.4 mM)/100 ml
methoxamine and quipazine, respectively, at day 133 postlesion.
1530
E. BRUSTEIN AND S. ROSSIGNOL
of locomotion in the adult cat. I. Treadmill walking. J. Neurophysiol. 76:
849 – 866, 1996.
JORDAN, L. M., MCCREA, D. A., STEEVES, J. D., AND MENZIES, J. E. Noradrenergic synapses aand effects of noradrenaline on interneurons in the
ventral horn of the cat spinal cord. Can. J. Physiol. Pharmacol. 55: 399 –
412, 1977.
KIEHN, O., HULTBORN, H., AND CONWAY, B. A. Spinal locomotor activity in
acutely spinalized cats induced by intrathecal application of noradrenaline.
Neurosci. Lett. 143: 243–246, 1992.
KUYPERS, H.G.J.M. Anatomy of the descending pathways. In: Handbook of
Physiology. The Nervous System. Sensory Processes. Bethesda, MD: Am.
Physiol. Soc., 1981, sect. 1, vol. III, p. 597– 665.
KUYPERS, H.G.J.M. AND MAISKY, V. A. Retrograde axonal transport of horseradish peroxydase from spinal cord to brain stem cell groups in the cat.
Neurosci. Lett. 1: 9 –14, 1975.
KUYPERS, H.G.J.M. AND MAISKY, V. A. Funicular trajectories of descending
brain stem pathways in cat. Brain Res. 136: 159 –165, 1977.
LANGER, S. Z. Presynaptic receptors and their role in the regulation of transmitter release. Br. J. Pharmacol. 60: 481– 497, 1977.
LEE, H. AND HECKMAN, C. J. Influence of voltage-sensitive dendritic conductances on bistable firing and effective synaptic current in cat spinal motoneurons in vivo. J. Neurophysiol. 76: 2107–2110, 1996.
LEE, R. H. AND HECKMAN, C. J. Characteristics of persistent inward currents
underlying dendritic plateau potentials in spinal motoneurons. Soc. Neurosci. Abstr. 23: 1302,1997.
MARSHALL, K. C. Catecholamines and their actions in the spinal cord. In:
Handbook of the Spinal Cord: Pharmacology, edited by R. A. Davidoff.
New York: Dekker, 1983, p. 275–328.
MARTIN, R. F., JORDAN, L. M., AND WILLIS, W. D. Differential projections of
cat medullary raphe neurons demonstrated by retrograde labelling following
spinal cord lesions. J. Comp. Neurol. 182: 77– 88, 1978.
MATSUYAMA, K. AND DREW, T. The organization of the projections from the
pericruciate cortex to the pontomedullary brainstem in the cat. A study using
the anterograde tracer, Phaseolus vulgaris leucoagglutinin. J. Comp. Neurol.
389: 617– 641, 1997.
MILLER, J. F., PAUL, K. D., LEE, R. H., RYMER, W. Z., AND HECKMAN, C. J.
Restoration of extensor excitability in the acute spinal cat by the 5HT2
agonist DOI. J. Neurophysiol. 75: 620 – 628, 1996.
MORI, S. Integration of posture and locomotion in acute decebrate cats and in
awake, freely moving cats. Prog. Neurobiol. 28: 161–195, 1987.
ORLOVSKY, G. N. Activity of vestibulospinal neurons during locomotion. Brain
Res. 46: 85–98, 1972.
PETRAS, J. M. Cortical, tectal and segmental fiber connections in the spinal
cord of the cat. Brain Res. 6: 275–324, 1967.
ROSSIGNOL, S. Neural control of stereotypic limb movements. In: Handbook of
Physiology. Exercise. Regulation and Integration of Multiple Systems, Bethesda, MD: Am. Physiol. Soc., 1996, sect. 12, p. 173–216.
ROSSIGNOL, S. AND BARBEAU, H. Pharmacology of locomotion: an account of
studies in spinal cats and spinal cord injured subjects. J. Am. Paraplegia
Soc. 16: 190 –196, 1993.
ROSSIGNOL, S., CHAU, C., BRUSTEIN, E., BELANGER, M., BARBEAU, H., AND
DREW, T. Locomotor capacities after complete and partial lesions of the
spinal cord. Acta Neurobiol. Exp. 56: 449 – 463, 1996.
SCHOMBURG, E. D. AND STEFFENS, H. Bistable characteristics of motoneuron
activity during DOPA induced fictive locomotion in spinal cats. Neurosci.
Res. 26: 47–56, 1996.
STEEVES, J. D., SCHMIDT, B. J., SKOVGAARD, B. J., AND JORDAN, L. M. Effect
of noradrenaline and 5-hydroxytryptamine depletion on locomotion in the
cat. Brain Res. 185: 349 –362, 1980.
STEVENS, R. T., APKARIAN, A. V., AND HODGE, C.J.J. Funicular course of
catecholamine fibers innervating the lumbar spinal cord of the cat. Brain
Res. 336: 243–251, 1985.
TATOR, C. H., DUNCAN, E. G., EDMONDS, V. E., LAPCZAK, L. I., AND ANDREWS,
D. F. Changes in epidemiology of acute spinal cord injury from 1947 to
1981. Surg. Neurol. 40: 207–215, 1993.
Downloaded from http://jn.physiology.org/ by 10.220.33.5 on June 18, 2017
BRUSTEIN, E., DE SYLVA, N., ROSSIGNOL, S., AND DREW, T. Modifications in
density and distribution of HRP-labelled cells in the brain stem and motor
cortex of adult cats after lesions to ventral and ventro-lateral spinal tracts.
Soc. Neurosci. Abstr. 23: 765, 1997.
BRUSTEIN, E. AND ROSSIGNOL, S. Recovery of locomotion after ventral and
ventrolateral spinal lesions in the cat. I. Deficits and adaptive mechanisms.
J. Neurophysiol. 80: 1245–1267, 1998.
CHAU, C., BARBEAU, H., AND ROSSIGNOL, S. Early locomotor training with
Clonidine in spinal cats. J. Neurophysiol. 79: 392– 409, 1998a.
CHAU, C., BARBEAU, H., AND ROSSIGNOL, S. The effects of intrathecal alpha-1
and alpha-2 noradrenergic agonists and noradrenaline on locomotion in
chronic spinal cats. J. Neurophysiol. 79: 2941–2963, 1998b.
CONWAY, B. A., HULTBORN, H., KIEHN, O., AND MINTZ, I. Plateau potentials in
alpha-motoneurones induced by intravenous injection of L-Dopa and
Clonidine in the spinal cat. J. Physiol. (Lond.) 405: 369 –384, 1988.
DIETZ, V., COLOMBO, G., JENSEN, L., AND BAUMGARTNER, L. Locomotor
capacity of spinal cord in paraplegic patients. Ann. Neurol. 37: 574 –582,
1995.
ERIKSSON, J., OLAUSSON, B., AND JANKOWSKA, E. Antispastic effects of L-dopa.
Exp. Brain Res. 111: 296 –304, 1996.
FORSSBERG, H. AND GRILLNER, S. The locomotion of the acute spinal cat
injected with Clonidine i.v. Brain Res. 50: 184 –186, 1973.
GIROUX, N., ALOYZ, R. S., ROSSIGNOL, S., AND READER, T. A. Serotonin 1a and
a1, a2-noradrenergic receptors in the spinal cord of spinalized cats. Soc.
Neurosci. Abstr. 21: 926, 1995.
GORSKA, T., BEM, T., MAJCZYNSKI, H., AND ZMYSLOWSKI, W. Unrestrained
walking in cats with partial spinal lesions. Brain Res. Bull. 32: 241–249,
1993a.
GORSKA, T., MAJCZYNSKI, H., BEM, T., AND ZMYSLOWSKI, W. Hindlimb swing,
stance and step relationships during unrestrained walking in cats with lateral
funicular lesion. Acta Neurobiol. Exp. 53: 133–142, 1993b.
GRILLNER, S. The role of muscle stiffness in meeting the changing postural and
locomotor requirements for force development by the ankle extensors. Acta
Physiol. Scand. 86: 92–108, 1972.
GRILLNER, S. Locomotion in the spinal cat. Adv. Behav. Biol. 7: 515–535,
1973.
GRILLNER, S. Control of locomotion in bipeds, tetrapods, and fish. In: Handbook of Physiology. The Nervous System. Motor Control. Bethesda: Am.
Physiol. Soc., 1981, sect. 1 vol. II, p. 1179 –1236.
GRILLNER, S. AND SHIK, M. L. On the descending control of the lumbosacral
spinal cord from the mesencephalic locomotor region. Acta Physiol. Scand.
87: 320 –333, 1973.
GRILLNER, S. AND ZANGGER, P. On the central generation of locomotion in the
low spinal cat. Exp. Brain Res. 34: 241–261, 1979.
GUERTIN, P., ANGEL, M., PERREAULT, M.-C., AND MCCREA, D. A. Ankle
extensor group I afferents excite extensors throughout the hindlimb during
MLR-evoked fictive locomotion in the cat. J. Physiol. (Lond.) 487: 197–209,
1995.
HELGREN, M. E. AND GOLDBERGER, M. E. The recovery of postural reflexes and
locomotion following low thoracic hemisection in adult cats involves compensation by undamaged primary afferent pathways. Exp. Neurol. 123:
17–34, 1993.
HOUNSGAARD, J., HULTBORN, H., JESPERSEN, J., 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.
HOWE, J. R., YAKSH, T. L., AND GO, V L. The effect of unilateral dorsal root
ganglionetomies or ventral rhizotomies on alpha2-adrenoreceptor binding
to, and the substance P, enkephalin, and neurotensin content of, the cat
lumbar spinal cord. Neuroscience 21: 385–394, 1987.
JANKOWSKA, E., HAMMAR, I., DJOUHRI, L., HEDEN, C., SZABO-LACKBERG, Z.,
AND YIN, X. K. Modulation of responses of four types of feline ascending
tract neurons by Serotonin and noradrenaline. Eur. J. Neurosci. 9: 1375–
1387, 1997.
JANKOWSKA, E., RIDDELL, J. S., SKOOG, B., AND NOGA, B. R. Gating of
transmission to motoneurones by stimuli applied in the locus coeruleus and
raphe nuclei of the cat. J. Physiol. (Lond.) 461: 705–722, 1993.
JIANG, W. AND DREW, T. Effects of bilateral lesions of the dorsolateral funiculi
and dorsal columns at the level of the low thoracic spinal cord on the control