| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Physiology, University of Manitoba, Winnipeg, R3E 0W3 Manitoba, Canada
 |
ABSTRACT |
|---|
 Abstract
 Introduction
Methods
Results
Discussion
References
|
|---|
Magnuson, David S. K. and Tammy C. Trinder. Locomotor rhythm evoked by ventrolateral funiculus stimulation in the neonatal rat spinal cord in vitro. J. Neurophysiol. 77: 200-206, 1997. Spinal cords from 2- to 8-day-old rats, maintained in vitro, were used to investigate the effects of discrete electrical stimuli applied to the ventrolateral funiculus (VLF) on motor neuron activity recorded from the lumbar ventral roots. Short trains of stimuli (1-3 s) delivered to one VLF in the low cervical region elicited rhythmic activity that persisted for up to 30 s. Responses consisted of short periods of activity (1-5 s) occurring simultaneously in the ipsilateral L5 and contralateral L3 ventral roots that alternated with similar bursts of activity in the ipsilateral L3 ventral root, a pattern consistent with locomotion. The rhythmicity of the ventral root responses to VLF stimulation was not affected by midsagittal sectioning of the preparation rostral to T10 and/or caudal to L4. Midsagittal sectioning of the lower thoracic or upper lumbar segments, however, disrupted the rhythmicity of the ventral root responses, leaving only long-duration simultaneous activation of the ipsilateral roots following VLF stimulus trains. The minimum lesion that effectively abolished the rhythmicity was one that divided only the L2 and L3 segments. In preparations rendered arrhythmic to VLF stimulation by an L2/L3 midsagittal lesion, rhythmicity could still be induced by N-methyl-D-aspartate (NMDA; 2-5 µM) and serotonin (5-HT; 20-50 µM), a drug combination commonly used to induce locomotor-like rhythmicity and air-stepping in vitro. Field potentials recorded following single stimuli delivered to the VLF revealed short-latency, large-amplitude responses in the ventral horn and intermediate gray both ipsilateral and contralateral to the stimulus site at T12 and L2. These observations suggest that 1) the discrete pathway under study may be an important descending locomotor command pathway and 2) this pathway has a strong bilateral projection in the lower thoracic and upper lumbar segments that is crucial for the initiation of VLF-induced rhythmic motor output. The induction of rhythmicity by NMDA/5-HT in an L2/L3-lesioned preparation suggests that these two rhythmogenic mechanisms may represent different levels within the circuitry that comprises the central pattern generator for locomotion. The rhythmic activity resulting from VLF stimulation is dependent on a bilateral projection that can be bypassed by the generalized excitation and subsequent rhythmicity that results from bath application of the NMDA/5-HT combination.
The initiation of locomotion in mammals is believed to involve the activation of specific brain stem and mesencephalic nuclei (see Jordan 1991 Sprague-Dawley rats aged 2-8 days were decapitated and eviscerated. The forelimbs and ribs were removed to within 0.5 cm of the spinal column. The hindlimbs and spinal column were quickly chilled to 4°C in oxygenated (95% O2-5% CO2) artificial cerebral spinal fluid (ACSF) that contained (in mM) 124.5 NaCl, 5.0 KCl, 26.0 NaHCO3, 1.0 NaH2PO4, 10.0 D-glucose, 2.0 MgSO4, and 2.0 CaCl2. The spinal cords were exposed ventrally, carefully dissected free, and placed in a 30°C recirculating incubating chamber filled with bubbled ACSF. Care was taken to retain, attached and undamaged, 0.5- to 1.0-cm lengths of lumbar ventral and dorsal roots. After In total, 51 preparations were included in the study. VLF-stimulation-induced rhythmic activity that alternated between ipsilateral and contralateral L3 ventral roots (as a minimum criterion for inclusion) was elicited from 41 preparations. Six of these successful preparations were used for field potential recording and eight were used for lesioning experiments. The remaining 27 rhythmic preparations were studied for reproducibility of response and to examine stimulation parameters including the positioning of electrodes. The stimulating electrodes were routinely placed on the VLF in the lower cervical/upper thoracic region of the isolated spinal cords.
Until recently, the terms "locomotor," "locomotor-like," and "rhythmic" activity have not been well defined for the neonatal rat spinal cord in vitro preparation. Observations by Cowley and Schmidt (1994a
INTRODUCTION
Abstract
Introduction
Methods
Results
Discussion
References
for review). The primary pathway is thought to originate in the mesencephalic locomotor region, passing through relays in the reticulospinal system of the lower brain stem and descending to the cervical and lumbar spinal cord in the ventrolateral funiculus (VLF) (Noga et al. 1991; Steeves and Jordan 1980; Yamaguchi 1986). Although a number of laboratories have described locomotor activity elicited by brain stem (Atsuta et al. 1988), mesencephalic (Skinner and Garcia-Rill 1984), or whole cord electrical stimulation (Iwahara et al. 1991), in vitro investigations into the central pattern generator for locomotion in the mammalian spinal cord have, to date, relied heavily on models utilizing N-methyl-D-aspartate (NMDA) and/or serotonin (5-HT) to induce locomotor-like rhythms (Cazalets et al. 1992; Kjærulff et al. 1994; Kudo and Yamada 1987).
, 1996). Complementary to that observation are those of Kjærulff and Kiehn, who found that incomplete L2/L3 transections, sparing only small portions of the most ventral white and gray matter, permit coordination between the rostral and caudal lumbar segments to be retained during drug-induced locomotor-like activity. They further demonstrated that when completely separated from the more rostral cord, the segments L3-L6 are capable of producing a coherent alternating right-left locomotor-like pattern in the presence of 5-HT and NMDA (Kjærulff and Kiehn 1994, 1995). So, although the debate is ongoing, these studies suggest that the pattern-generating network for the hindlimbs, in the neonatal rat at least, is located throughout the lumbar enlargement but is dominated by neurons in the more rostral segments (L1 and L2) when a longitudinally contiguous preparation is used. The distribution of the locomotor pattern generated in the rostral segments to the rest of the lumbar enlargement may occur via the most ventral portion of the cord.
) and cfos induction following treadmill activity (in vivo) (Dai et al. 1990) was used have suggested that neurons located in laminae VII and X of the lumbar segments are involved in those motor activities. We have recently described rhythmic fluctuations in membrane potential exhibited by lumbar lamina VII interneurons recorded intracellularly in a lumbar midsagittally sectioned preparation of neonatal rat spinal cord in vitro (midsagittal section of lumbar enlargement with 1 lumbar hemicord removed, thoracic cord intact) (MacLean et al. 1995). We found that ~80% of the L2-L5 lamina VII interneurons impaled with the use of sharp microelectrodes exhibited "drive potentials" in phase with 5-HT/NMDA-induced rhythmic ventral root activity. The amplitudes of these fluctuations in potential were modest, however, bearing little resemblance to drive potentials recorded from motoneurons in the cat (for example Shefchyk and Jordan 1985) or rat (for example Schmidt 1994), or from interneurons in the mudpuppy (Wheatley et al. 1994). The possibility exists that a proportion of these neurons were activated unspecifically, and exhibited fluctuations in membrane potential through a process akin to common excitation. Because similar experiments performed in a drug-free model of locomotion would suffer less from this problem, we set out to develop a preparation that might allow pattern generation and descending locomotor command pathways to be studied without the need for drugs to induce rhythmic motor activity. Thus the present study examines the responses in lumbar ventral roots to stimulation of a specific descending pathway in VLF of the isolated neonatal rat spinal cord. The pathway is identified by location and by the high dependence the ventral root responses have on the stimulus frequency (Magnuson et al. 1995).
METHODS
Abstract
Introduction
Methods
Results
Discussion
References
30 min, the cords were transferred to a Sylgard-bottomed recording chamber continuously perfused (1.5-2.0 ml/min) with bubbled ACSF at room temperature. The cords were pinned out, ventral surface uppermost, with the use of 0.1-mm-diam insect pins placed in the cervical, midthoracic, and sacral regions of the isolated cord.
(tip diameter 2-5 µm; filled with ACSF) lowered onto the ventrolateral white matter with the use of a Newport 461 three-dimensional mechanical micromanipulator. No suction was applied to the stimulating electrode, and stimuli were delivered by the A-M Systems constant-current 1200 stimulator. Stimulus intensities were determined as multiples of the ipsilateral L3 ventral root threshold (range 5-30 µA). Rhythmic ventral root activity was elicited most consistently by stimuli lasting 0.5 ms, delivered at 50 Hz for 0.5-3.0 s, at 2-5 times threshold. The stimulating electrode was placed on the caudal region of the cervical enlargement, approximately one third of the way between the line formed by the exit points of the ventral roots and the line formed by the entry points of the dorsal roots (at ~4 o'clock on a cross section). The stimulating electrode was placed so as not to dimple the cord more than slightly, and small adjustments in position were commonly required to find the optimal stimulation site for each individual cord. Lowering the stimulating electrode firmly onto the white matter or inserting the electrode into the white matter were not successful strategies. In otherwise uncompromised preparations, our success rate in eliciting rhythmic locomotor-like activity with VLF stimulation was ~80%.
(filled with ACSF). Averages of three to six sweeps were made following single 0.5-ms stimuli delivered at 0.1 Hz and signals were high-pass filtered at 0.1 Hz. The field potential electrodes were lowered into the cord almost vertically with the use of a Newport MX-510 hydraulic micromanipulator mounted on a Newport 461 three-dimensional mechanical micromanipulator. The depth of penetration of the field potential electrode tip was determined by the cylinder graduations on the MX-510 hydraulic manipulator. Care was taken to establish the depth of penetration when no dimpling of the cord was visible when viewed through the dissecting microscope. In some instances this required that the electrode be moved slowly through the cord beyond the desired depth and then be withdrawn slightly before making any records at a particular site. As a routine, records were made starting from the ventral surface.
RESULTS
Abstract
Introduction
Methods
Results
Discussion
References
). Stimulus durations of 0.3-1.0 ms had ventral root thresholds of 5-10 µA. At stimulus durations of 0.5 ms, rhythmic activity could be elicited with as little as 10 µA when the stimulus trains were 20 Hz in frequency and 5 s in duration. Only rarely were we able to initiate and maintain some level of rhythmicity with the use of low-frequency continuous stimulation of 1-10 Hz; the rhythmic activity generally ceased after only one or two cycles. In apparent contrast, short trains of higher-frequency stimulation (0.5-3.0 s; 50 Hz, 0.5 ms, 10-80 µA) could be used to elicit fairly uniform responses every 5-10 min for 2 h or more in some preparations. Although there was an obvious relationship between stimulus parameters and quantitative aspects of the rhythmic response (amplitude of burst, frequency of firing within a burst, and number of rhythm cycles), this relationship held only up to a certain threshold; exceeding that threshold yielded no further increases in the rhythmic responses and increases in background activity were common. For the majority of preparations (~20 of 27) this threshold was 
50 Hz and 50 µA.
View larger version (53K):
[in a new window]
FIG. 1.
A: rhythmic responses to a short-train stimulation of the low cervical ventrolateral funiculus (VLF) involving the contralateral L3 ventral root in addition to the ipsilateral L3 and L5 ventral roots. B: another example of rhythmic activity in response to a short-train stimulation of the VLF, this time showing alternation between ipsilateral and contralateral L3 ventral roots. C: 2 rhythmic responses from the same preparation, 1 following a 3-s, 50-Hz train of stimuli delivered to the VLF (left) and the 2nd following mild abrasion of the stimulus site with a 0.1-mm-diam insect pin (right).
View larger version (28K):
[in a new window]
FIG. 2.
Diagrams representative of lesions created manually with the use of 0.1-mm insect pins. Following the lesions shown on the left, low-cervical-VLF-stimulation-elicited rhythmic responses were unaffected. The lesions shown on the right completely abolished VLF-elicited rhythmic activity. Intermediate lesions involving the lower thoracic or upper lumbar segments leaving intact L2 or L3 resulted in a disruption but not an abolishment of rhythmic responses.
View larger version (30K):
[in a new window]
FIG. 3.
A: 1 example where a minimum effective lesion (L2 and L3) abolished rhythmic response to train stimulation of the VLF. B: in same preparation, the L2/L3-lesioned cord still responded rhythmically to the application of serotonin (5-HT)/N-methyl-D-aspartate (NMDA) in the superfusate.
; Magnuson et al. 1995).
View larger version (57K):
[in a new window]
FIG. 4.
Representative field potentials from the T12, L2, and L4 spinal cord segments following single VLF stimuli. Records are shown in relation to a diagram showing the approximate relative location of the motor nuclei in the various spinal cord segments represented. These records suggest that VLF inputs are greatest at T12 and L2, with strong bilateral projections at L2 and, in some preparations, at T12 and L4 (see Table 1 for averaged responses).
View this table:
TABLE 1.
Amplitudes and latencies of field potentials
View larger version (10K):
[in a new window]
FIG. 5.
Neuronal responses recorded extracellularly following VLF stimulation were either short in latency (A), of intermediate latency (B), or long in latency (C) compared with the latency of the responses in the L3 ventral root ipsilateral to the stimulus site.
DISCUSSION
Abstract
Introduction
Methods
Results
Discussion
References
,b), using simultaneously recorded ventral root and ankle flexor and extensor electroneurograms, suggest that drug-induced rhythmic activity recorded from ventral roots cannot always be interpreted as locomotor or locomotor-like activity. However, these authors do not give examples where alternating rhythmic activity recorded in either L2 or L3 and L5 ventral roots was seen as something other than a locomotor-like pattern of activity in flexor and extensor electroneurograms, respectively (Cowley and Schmidt 1994a,b). In addition, the recent report by Kiehn and Kjærulff (1996) demonstrates that L5 ventral root bursts correspond to extensor electromyogram activity during drug-induced locomotor patterns. In the present study we chose to record from L3 and L5 ventral roots, which contain predominantly flexor and extensor motor axons, respectively. Although the absence of identified flexor and extensor electroneurograms makes it impossible to positively assign stepping phases to the recorded ventral root activity, the findings of Cowley and Schmidt and Kiehn and Kjærulff outlined above suggest that the particular pattern of alternating activity in L3 and L5 ventral roots is unlikely to reflect something other than a locomotor pattern of the postnatal day 2-8 rat pup.
). In this preparation, ~80% of intermediate laminae interneurons in L2-L5 were found to exhibit rhythmic fluctuations in membrane potential that were phase linked to rhythmic activity in nearby ventral roots. We also demonstrated recently that relatively low-intensity, high-frequency stimulation of the cervical VLF in the spinal cord-hindlimb preparation (intact spinal cord) can elicit long bouts of rhythmic hindlimb movements that resemble locomotion (Magnuson et al. 1995). When we attempted to induce rhythmic activity by VLF stimulation in the lumbar midsagittally sectioned spinal cord, however, we were unsuccessful, observing only long-duration tonic coactivation of lumbar ventral roots on the intact hemicord. Thus we faced a conundrum whereby both VLF stimulation and 5-HT/NMDA could elicit rhythmic locomotor-like activity in the intact cord, but only the 5-HT/NMDA was successful in the lumbar midsagittally sectioned preparation. This conundrum was solved, however, when we midsagittally divided the upper lumbar segments in spinal cord-hindlimb preparations that had already demonstrated locomotor-like activity in response to stimulation of the VLF. After the midsagittal sectioning of the L2 and L3 segments, VLF stimulation elicited only cocontractions in the flexor and extensor muscles of the ipsilateral hindlimb; these responses appeared disorganized and uncoordinated with occasional rapid shaking and twitching (small spasmlike movements; unpublished observations). These same preparations displayed robust locomotor-like activity when treated with 5-HT/NMDA. These observations indicated that the initiation of rhythmic activity by VLF stimulation was dependent on fibers that crossed the midline in the upper lumbar segments. Drug-induced rhythmicity is not dependent on this pathway, but can be influenced by it; this we demonstrated by stimulating the VLF during drug-induced rhythmic activity, observing that VLF stimulation could perturb the ongoing rhythm (Magnuson et al. 1995).
with the use of similar recording and filtering paradigms. Their field potentials were elicited by stimulation of the mesencephalic locomotor region located just dorsal to the brachium conjunctivum in the vicinity of the cuneiform nucleus. Although the locations of the cell bodies giving rise to the VLF fibers we stimulated are not known, the fibers themselves are located in a region previously shown to carry the descending output from the mesencephalic locomotor region in the cat (Noga et al. 1991). The occasional recording from a presumed neuronal cell body following single stimuli (Fig. 5) further supports the contention that the stimulated fibers descend to the lower thoracic and upper lumbar segments in the VLF and enter the ventral horn both ipsilaterally and contralaterally. The short-duration, large-amplitude extracellular potentials did not follow high-frequency (>10 Hz) stimulation, but were consistently elicited by low-frequency stimuli delivered at 5 times the ventral root threshold. The majority of these responses were found contralateral to the stimulus site, had short latencies, and may represent monosynaptic inputs onto intermediate laminae neurons. This type of response was perhaps not reported by Noga et al. (1995), because they consistently recorded their field potentials ipsilateral to the mesencephalic locomotor region stimulus site, and they made no recordings rostral to L3.
. Functionally, one could speculate that the bilateral symmetry of output from one mesencephalic locomotor region might assist in making the transition from standing to walking, which requires simultaneous changes in motor nuclei on both sides of the spinal cord, a smooth one.
| |
ACKNOWLEDGEMENTS |
|---|
We thank C. Gibbs for excellent technical assistance and N. Braver for helpful editorial comments.
This work was supported by a grant from the Medical Research Council of Canada to D. S. K. Magnuson.
| |
FOOTNOTES |
|---|
Present address and address for reprint requests: D. S. K. Magnuson, Dept. of Neurological Surgery, 210 East Gray St., Suite 1102, University of Louisville, Louisville, KY 40292.
Received 30 May 1996; accepted in final form 28 August 1996.
| |
REFERENCES |
|---|
|
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Rossignol, G. Barriere, O. Alluin, and A. Frigon Re-expression of Locomotor Function After Partial Spinal Cord Injury Physiology, April 1, 2009; 24(2): 127 - 139. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. F. Samara and S. N. Currie Location of Spinal Cord Pathways That Control Hindlimb Movement Amplitude and Interlimb Coordination During Voluntary Swimming in Turtles J Neurophysiol, April 1, 2008; 99(4): 1953 - 1968. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. C. Cowley, E. Zaporozhets, and B. J. Schmidt Propriospinal neurons are sufficient for bulbospinal transmission of the locomotor command signal in the neonatal rat spinal cord J. Physiol., March 15, 2008; 586(6): 1623 - 1635. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Falgairolle and J.-R. Cazalets Metachronal coupling between spinal neuronal networks during locomotor activity in newborn rat J. Physiol., April 1, 2007; 580(1): 87 - 102. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Guevremont, J. A. Norton, and V. K. Mushahwar Physiologically Based Controller for Generating Overground Locomotion Using Functional Electrical Stimulation J Neurophysiol, March 1, 2007; 97(3): 2499 - 2510. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Zaporozhets, K. C. Cowley, and B. J. Schmidt Propriospinal neurons contribute to bulbospinal transmission of the locomotor command signal in the neonatal rat spinal cord J. Physiol., April 15, 2006; 572(2): 443 - 458. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. P. Carlin, Y. Dai, and L. M. Jordan Cholinergic and Serotonergic Excitation of Ascending Commissural Neurons in the Thoraco-Lumbar Spinal Cord of the Neonatal Mouse J Neurophysiol, February 1, 2006; 95(2): 1278 - 1284. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. R. Noga, D. J. Kriellaars, R. M. Brownstone, and L. M. Jordan Mechanism for Activation of Locomotor Centers in the Spinal Cord by Stimulation of the Mesencephalic Locomotor Region J Neurophysiol, September 1, 2003; 90(3): 1464 - 1478. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Strauss and A. Lev-Tov Neural Pathways Between Sacrocaudal Afferents and Lumbar Pattern Generators in Neonatal Rats J Neurophysiol, February 1, 2003; 89(2): 773 - 784. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. N. Loy, D. S. K. Magnuson, Y. P. Zhang, S. M. Onifer, M. D. Mills, Q.-l. Cao, J. B. Darnall, L. C. Fajardo, D. A. Burke, and S. R. Whittemore Functional Redundancy of Ventral Spinal Locomotor Pathways J. Neurosci., January 1, 2002; 22(1): 315 - 323. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. MacKay-Lyons Central Pattern Generation of Locomotion: A Review of the Evidence Physical Therapy, January 1, 2002; 82(1): 69 - 83. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Garraway and S. Hochman Serotonin Increases the Incidence of Primary Afferent-Evoked Long-Term Depression in Rat Deep Dorsal Horn Neurons J Neurophysiol, May 1, 2001; 85(5): 1864 - 1872. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Marchetti, M. Beato, and A. Nistri Alternating rhythmic activity induced by dorsal root stimulation in the neonatal rat spinal cord in vitro J. Physiol., January 1, 2001; 530(1): 105 - 112. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. N. Currie and G. G. Gonsalves Reciprocal Interactions in the Turtle Hindlimb Enlargement Contribute to Scratch Rhythmogenesis J Neurophysiol, June 1, 1999; 81(6): 2977 - 2987. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
