| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Cell Biology and Neuroscience, University of California–Riverside, Riverside, California
Submitted 29 September 2007; accepted in final form 11 February 2008
 |
ABSTRACT |
|---|
|
|
|
INTRODUCTION |
|---|
|
), or fictive forward swim motor patterns in contralateral forelimb and hindlimb nerves (Currie 2003). The locations of the spinal pathways that carry these descending locomotor commands to motor pattern generating circuitry in the spinal limb enlargements in turtles are currently unknown. However, focal electrical stimulation of the dorsolateral funiculus (DLF) just caudal to the forelimb enlargement (spinal segments D2–D3) has evoked forward swim movements of the contralateral hindlimb in intact and low-spinal turtles (Lennard 1985; Lennard and Stein 1977). Electrical stimulation of sites outside of the DLF failed to evoke swimming movements. In addition, DLF stimulation at the first cervical segment in high-spinal turtles typically initiated coordinated forward swimming in the contralateral forelimb and hindlimb (Stein 1978). Fictive forward swimming has also been elicited in low-spinal-immobilized (Berkowitz 2002; Juranek and Currie 2000) and decerebrate-immobilized turtles (Currie 2003).
Studies with other vertebrates have sought the location of locomotor command pathways in the spinal cord by mapping the spinal white matter with focal electrical stimulation ("sufficiency studies") and lesioning the spinal white matter in preparations that locomoted spontaneously or in response to electrical/chemical brain stimulation ("necessity studies"). In cats, experiments using cervical spinal lesions and electrical stimulation of the mesencephalic locomotor region (MLR) suggested that the ventrolateral funiculus (VLF) contains fibers that initiate stepping in the ipsilateral forelimb (Yamaguchi 1986) or hindlimb (Noga et al. 1991; Steeves and Jordan 1980), although the DLF also appeared to play a role (Noga et al. 1991; Yamaguchi 1986). Additionally, reversible cooling of the VLF bilaterally blocked locomotion induced by electrical stimulation of the pontomedullary locomotor region (PLR) (Noga et al. 1991). In stingrays, electrical stimulation of the DLF produces swimlike activity in the contralateral fin of moving preparations, but not in immobilized preparations (Williams et al. 1984). However, lesioning the DLF does not impair brain stem stimulation-evoked swim movements (Livingston and Leonard 1990). Additionally, stingrays were unable to swim spontaneously only following bilateral lesions of the intermediate lateral white matter (Williams et al. 1984). Evidence in lampreys also indicates a relatively diffuse distribution of descending swim pathways because sparing either the VLF or the DLF allowed swim behavior, but bilaterally lesioning the lateral white matter altogether did not (McClellan 1988).
Turtles are especially well suited for studies on interlimb coordination since the slight mechanical coupling between forelimbs and hindlimbs, as caused by water currents and internal viscera during carapace-restrained swimming, is minimal compared to the mechanical coupling characteristic of overground stepping in other preparations. During voluntary forward swimming, fresh-water turtles typically display a 1:1 coordination between all four limbs, including right–left alternation in the forelimbs and in the hindlimbs, out-of-phase coordination of ipsilateral forelimb–hindlimb pairs, and nearly in-phase coordination between diagonal limbs (Davenport et al. 1984; Earhart and Stein 2000; Field and Stein 1997; Zug 1971). Although considerable evidence implicates the DLF in turtle locomotor commands, less is known about the location of pathways that mediate interlimb coordination. In experiments by Stein (1978), the contralateral forelimb and hindlimb movements elicited by DLF stimulation in high-spinal turtles usually exhibited 1:1, out-of-phase coordination, showing that intact supraspinal circuitry was not needed for proper forelimb–hindlimb coordination. Data in chicks (Bekoff et al. 1989; Jacobson and Hollyday 1982) and cats (Miller and van der Meche 1976) also demonstrate that proper interlimb coordination can be observed in the absence of intact supraspinal centers. Furthermore, bilateral deafferentation of the turtle hindlimb enlargement showed that movement-related sensory feedback was not necessary for normal interlimb coordination during voluntary swimming (Samara and Currie 2007). Other evidence indicated that forelimb–hindlimb coordination did not require movement-related sensory feedback, since electrical stimulation in the brain stem of decerebrate-immobilized preparations evoked coordinated out-of-phase fictive forward swimming in forelimb and hindlimb nerves (Currie 2003). Isolated rat spinal cords produce fictive motor output resembling that seen during actual locomotion in response to chemical stimulation (Juvin et al. 2005). In low-spinal, immobilized cats, fictive forelimb–hindlimb motor output with interlimb coordination that resembles actual locomotion is observed in response to chemical and electrical spinal cord stimulation (Grillner and Zangger 1979). These results suggest that the spinal cord itself contains sufficient circuitry for 1:1 interlimb coordination. Furthermore, propriospinal fibers connecting the forelimb and hindlimb enlargements have been demonstrated in the ventral and/or ventrolateral white matter in turtles (Kusuma and ten Donkelaar 1980), cats (Giovanelli Barilari and Kuypers 1969), and rats (Reed et al. 2006), indicating likely locations of axons contributing to interlimb phase control. Other studies in cats (Bem et al. 1995; Brustein and Rossignol 1998; Gorska et al. 1993a,b, 1996; Jiang and Drew 1996; Kato 1992; Zmyslowski et al. 1993) and rats (Loy et al. 2002a,b; Schucht et al. 2002) assessed the effects of spinal lesions on forelimb-hindlimb coordination, suggesting that both the DLF and VLF contained axons contributing to interlimb coupling.
In the present experiments, we performed unilateral and bilateral lesions of the D2–D3 spinal cord in otherwise intact, carapace-restrained turtles to localize the spinal cord pathways that regulate the amplitude of voluntary hindlimb swimming movements and coordinate those movements with other limbs. Our results indicate that locomotor command pathways that initiate and control the amplitude of voluntary hindlimb swimming are concentrated in the dorsolateral funiculus (DLF) at the midbody D2–D3 level and cross over to affect contralateral hindlimb movements. We also show that forelimb–hindlimb coordination is maintained by spinal pathways that are largely confined to the ventrolateral funiculus (VLF) and mediate phase coupling of ipsilateral limbs. A preliminary report of these data was presented in abstract form (Samara and Currie 2004).
|
|
METHODS |
|---|
|
Spinal cord exposure
Prior to surgery, red-eared slider turtles, Trachemys scripta elegans (n = 41) with plastron lengths from 12 to 17 cm, were placed in crushed ice for 
2 h to induce hypothermic anesthesia (Lennard and Stein 1977). During all surgical procedures, turtles remained partially submerged in ice. The D2 and D3 segments of spinal cord, just posterior to the forelimb enlargement, were exposed by dorsal laminectomy and then covered with saline-soaked gel foam. Dental utility wax (Heraeus Kulzer, South Bend, IN) was used to cover the laminectomy region and glued to the surrounding carapace with Permabond adhesive until a spinal lesion would be made at a later time, following prelesion swim trials.
Movement recording
We held turtles suspended just below the surface in a water-filled Plexiglas tank [40 x 35 x 16 cm (length x width x height)] via a band clamp around the middle of the carapace. Voluntary swimming movements were videotaped from below in these carapace-restrained animals with a horizontally mounted digital video camera (Canon Optura 20 mini-DV), which was aimed at a 45° mirror beneath the clear bottom of the tank (see apparatus diagram in Samara and Currie 2007). Brightly colored markers (3-mm beads) were attached to the skin at the wrists (h and i in Fig. 1), knees (f and g), and ankles (k and l) on both sides. We tracked the movement of these markers relative to fixed reference points on the plastron, marked by dots of white typo correction fluid (a–e). Swim episodes were videotaped at a 30-Hz frame rate and 1/250-s shutter speed, then reviewed and annotated with a Sony GV-1000 Video Walkman. Marker positions were manually digitized frame by frame in selected swim episodes, using Datapac 2K2 software (Run Technologies, Mission Viejo, CA). Datapac 2K2 was used to measure changes in forelimb and hip angle over time and to calculate amplitude, interlimb phase, and cycle period from these swim episodes. Hip angle was defined as the angle between the thigh line (d–f on the right or e–g on the left in Fig. 1) and a line parallel to the ventral midline (a–c) with its origin at the hip joint (dashed line in Fig. 1). The hip joint (acetabulum) moves very little in the horizontal plane, even during vigorous locomotion (Walker 1971; PSG Stein, RF Samara, and SN Currie, unpublished observations). Forelimb angle was defined as 
abh on the right and abi on the left, and was measured relative to a stationary reference point on the plastron midline (b), which corresponded to the anteroposterior level of right and left pivot points on the pectoral girdle (Samara and Currie 2007; Walker 1971). "Forelimb angle" was used rather than "shoulder angle" because the shoulder joint (glenoid cavity) moves through a wide arc when the pectoral girdle rotates (Walker 1971), so its position cannot be determined without X-ray cinematography. Head angle was defined as abj, where an angle of 0° indicated that the head was pointed straight forward and movement of the head to the right or left resulted in positive or negative angles, respectively.
|
After acquiring a sufficient number of prelesion swim cycles, turtles were returned to ice before spinal lesions were made between the D2 and D3 dorsal roots. Short (2-mm-long) pieces of cord were removed from this region, rather than making single transverse cuts, so that the lesions would not close up and would be more easily and accurately traced during histological examination. The dental wax and gel foam that covered the D2–D3 exposure were removed and a microscalpel (broken from a razor blade) was used to make a 2-mm longitudinal cut in the cord. For lateral hemisections, this incision was made in the midsagittal plane along the visible dorsal median sulcus and spanned the entire cord. For dorsal and ventral hemisections, the incision was made through the midhorizontal plane after twisting the cord 90°. The narrowness of the midbody cord (width 1.5–2 mm), as well as the long distance between D2 and D3 spinal roots (12–15 mm), meant that the cord in this region could be twisted very easily with little resistance. We also cut the D2 and D3 spinal roots to further reduce any strain on the cord during twisting and to render the animal insensitive to the region of the D2–D3 cord exposure. The anterior and posterior margins of the longitudinal incision were then cut with iridectomy scissors and the 2-mm-thick section of cord was removed. Other types of lesions, either superficial or cord-spanning, were made using variations of this technique. The exposed region of cord was then re-covered with gel foam and dental wax before the turtle was returned to the water and allowed to warm to room temperature for 1 h before postlesion movement recordings.
Histology
At the conclusion of each experiment, turtles were returned to ice for 1 h to reinduce hypothermic anesthesia. The region of D2–D3 cord containing the lesion was removed together with several millimeters of intact cord anterior and posterior to the lesion and pinned out straight and untwisted by its loosely attached arachnoid membrane in a dish that contained 4% paraformaldehyde (refrigerated at 4°C) for approximately 1 wk. Some particularly extensive lesions that left only one or two thin bridges of tissue spanning the lesion site (e.g., R76 in Fig. 2A, R58 in Fig. 6) required special care when being removed from the animal and pinned out for fixation. Turtles were euthanized by freezing following cord removal. Prior to sectioning, the outer surface of the fixed cord was painted with Janus Green (Matheson, Cincinnati, OH) along its right side, so that we could later identify the right and left sides of each section. The fixed cord was embedded in agar, then the entire anteroposterior extent of the lesioned area and neighboring intact regions were sectioned at 80 µm in the transverse plane and stained with 0.1% cresyl echt violet (CellPoint Scientific, Rockville, MD), coverslipped, then visualized with a microprojector (Ken-A-Vision, Kansas City, MO) the following day. An intact section of cord immediately adjacent to the lesion site was traced by hand. The section containing the most extensive damage was then superimposed on the intact tracing. Using landmarks of the gray matter, the lesion boundary was traced onto the intact section. If a lesion removed all of the gray matter, the lesioned section was oriented such that the outside curvature matched the intact section as closely as possible. The resulting diagrams are shown in Fig. 2.
|
|
Data analysis
We quantified amplitude, interlimb phase, vector length, and cycle period in 40 of 41 turtles. Due to limited postlesion swim movements, we could not quantitatively analyze data from experiment R58. For all other turtles, analysis was performed on 25 forelimb swim cycles and their associated hip cycles, which were often lower in number postlesion. To be considered forward swimming, an episode had to exhibit 1) out-of-phase 1:1 movements between contralateral forelimbs with cycle periods no greater than 1.5 s and 2) head angles no greater than ±10° off-center to the right or left, which indicated that the turtle was swimming forward and not turning. The same cycles were used for amplitude, interlimb phase, and cycle period analyses.
We calculated peak-to-peak amplitude (±SD) in Datapac 2K2 by subtracting the minimum hip angle from the maximum hip angle for each cycle. Because our angle measurements were two-dimensional (2-D), limb movements outside of the horizontal plane would cause an inaccurate depiction of their amplitude. To establish that limb movements in a given episode were close to the horizontal plane, we obtained the maximum thigh length measurement from video records when the thigh was horizontal. We then calculated the ratio of each swim episode's minimum thigh length relative to the maximum length. This was done for the right and left hips from the 252 episodes we analyzed for a total of 504 ratio calculations. In the majority of cases (340 of 504), the minimum thigh length stayed within 80% of the maximum. In the remaining episodes, minimum thigh length was within 65% of the maximum. Since this was not true for the forelimbs (segments b–h and b–i) or the shanks (segments f–k and g–l), we did not quantify the amplitudes of forelimb or knee angles in our analyses. Postlesion hip movement amplitudes were normalized to mean prelesion amplitudes. We grouped amplitude data from different turtles based on whether lesions bilaterally spared, bilaterally damaged, or unilaterally damaged the DLF (Fig. 5). For each experiment in Fig. 5, postlesion hip movement amplitudes were normalized to mean prelesion amplitudes. Mean normalized amplitudes were then pooled across experiments within a particular group (e.g., "bilaterally spared DLF") to obtain the mean (±SD) of the mean amplitudes for that group. The Mann–Whitney U test (Siegel 1956) was used to determine whether normalized postlesion amplitudes were significantly different from prelesion controls.
|
; Field and Stein 1997); phase values were calculated with Datapac 2K2. One limb was selected as "referent" (RF or RH) and the other as "target" (Figs. 7 and 8; Tables 1–3). The onsets of referent-limb extension were defined by phase values of 0.0 and 1.0. The offsets of referent extension were defined by a phase value of 0.5 (see Field and Stein 1997). Phase data were imported into Oriana 2.0 (Kovach Computing Services, Anglesey, Wales, UK) to obtain circular statistics, which are appropriate for cyclical events (Batschelet 1981; Mardia and Jupp 2000; Zar 1999). The angle and length of the mean vector were calculated using standard trigonometric functions. The angle of the mean vector (µ) represents the average phase value on a circular scale ranging between 0.0 and 1.0. The length of the mean vector (r) indicates the directional concentration of data points around the mean vector angle. For Fig. 9, mean vector lengths were pooled across experiments within a particular group (e.g., "bilaterally spared VLF") to obtain the mean (±SD) of the mean vector lengths for that group. We used Rao's spacing test (Batschelet 1981; Levitin and Russell 1999; Mardia and Jupp 2000; Zar 1999) to test the null hypothesis that interlimb phase data were uniformly distributed; significant P values (<0.05) indicated that the null hypothesis was rejected, i.e., that the data were nonuniform and clustered around one or more phase values. Rao's spacing test is a powerful statistic that is applicable to circular data and can discriminate between uniform (random) and clustered distributions, regardless of whether the distributions are unimodal or multimodal (Batschelet 1981; Levitin and Russell 1999; Zar 1999). The Rayleigh test, on the other hand, can discriminate only between uniform and unimodal-clustered distributions; bimodal and multimodal distributions can test as not significantly different from uniform. Since a small subset of our postlesion coordination data appeared to be bimodal (2:1 coupling), we used the Rao statistic to discriminate this "relative coordination" (von Holst 1973) from noncoordination (uniformly distributed phase values). We used the Watson–Williams test (Batschelet 1981; Zar 1999) to determine whether mean postlesion phase values were significantly different from prelesion.
|
|
|
|
|
) was used to determine whether pre- and postlesion mean cycle periods were significantly different.
|
|
|
RESULTS |
|---|
|
Prior to spinal lesions, turtles exhibited large-amplitude, bilaterally symmetrical movements of all four limbs during vigorous forward swimming. See Data analysis (under METHODS) for the criteria we used to define forward swimming. Generally, the right and left forelimbs moved similar distances during each cycle, as did the right and left hips (Figs. 3 and 4). However, only hip amplitudes were quantified because they stayed close to a horizontal plane and could be quantified in 2-D video with some accuracy. This was not true for forelimb movements (see Data analysis).
|
|
; Lennard and Stein 1977) or fictive swimlike activity in contralateral hindlimb nerves (Berkowitz 2002; Juranek and Currie 2000). Based on those findings, we categorized lesions by the amount of DLF damage when analyzing their effects on the amplitude of hip swimming movements. Lesions that bilaterally spared the DLF (n = 5; all experiments in Fig. 2, A and B) produced no visible impairment of hip movement amplitudes in individual experiments (Figs. 3D and 4, B and D) and no statistically significant deficit in pooled amplitude data from the entire "DLF bilaterally spared" group (Fig. 5A). This included lesions that severed nearly all medial white matter (experiments R76 and R77, Figs. 2A and 4B). When the DLF was bilaterally damaged (n = 16; all experiments in Fig. 2, F–H), the most common effect was a reduction in right and left hip movement amplitudes (Figs. 3C and 4C), which was significant in data pooled from the "DLF bilaterally damaged" group (Fig. 5A). Five turtles had hip movements completely abolished on both sides by their lesions (experiments R41, R42, R66, R69, and R72 in Fig. 2G). These five experiments were excluded from the quantitative analyses of pooled data in Figs. 5 and 9. A sixth turtle (experiment R40 in Fig. 2G) lost all hip movement on the right side, but retained very low amplitude hip movements on the left side. Note that all six turtles had lesions that disrupted a significant portion of the DLF and VLF on both sides, suggesting that although swim-activating axons may be most concentrated in the DLF at the D2–D3 level (explaining why electrical stimulation in this region elicits hindlimb swimming), their distribution may be somewhat diffuse and extend into the VLF. Finally, we consistently found that unilateral damage of the DLF (n = 19; all experiments in Fig. 2, C–E) impaired movement amplitudes of the contralateral hip more severely than the ipsilateral hip (Figs. 3B and 5B). In the experimental group that received left DLF damage, the right (contralateral) hip showed more severe amplitude deficits than the left (ipsilateral) hip. In contrast, the group receiving right DLF damage displayed more severe deficits in left hip amplitude than right hip amplitude (Fig. 5B). In one experiment (experiment R56 in Fig. 2C), the unilateral damage was limited mainly to the dorsal funiculus (DF) above and medial to the dorsal horn and included little of the DLF; this animal exhibited no significant change in hip movement amplitude on either side, suggesting that it was damage to DLF rather than DF axons that impaired hip amplitudes. Pre- and postlesion forelimb movements were similar in amplitude for all experiments, indicating that reduced postlesion hip movements were not due to a general postsurgical sluggishness. The turtle shown in Fig. 6 (experiment R58) had the most extensive spinal lesion in our study, with only a small part of the right-side lateral white matter remaining intact, primarily composed of DLF. This experiment provided an especially compelling case for the activation of voluntary hindlimb swimming by crossed descending commands, since only the left hindlimb, contralateral to the intact white matter, displayed swimming movements. The ipsilateral right-side hindlimb remained completely immobile. Note that the slow left-side hindlimb swim movements occurred during accelerations in the forelimb swim frequency, suggesting that both were driven by periods of heightened descending drive. Because of the very small number of hindlimb swim cycles elicited in this animal, it was not included in quantitative analyses.
Interlimb coordination
PHASE ANALYSIS. Prelesion swim movements were out of phase between contralateral forelimbs, contralateral hips, and ipsilateral limbs and nearly in-phase between diagonal limbs (Fig. 7A). The mean pre- and postlesion interlimb phase values for each experiment are given in Tables 1–3. All prelesion phase distributions were tightly clustered near their respective means (P < 0.05, Rao's spacing test). Prelesion pooled mean vector lengths, which indicated how strongly phase coupled the limbs were on a scale from 0 to 1 (0 = uncoupled; 1 = strongly coupled), ranged from 0.92 to 0.95 for right–left forelimb coordination (RF-LF), right hip–right forelimb coordination (RH-RF), left hip–right forelimb coordination (LH-RF), and left–right hip coordination (LH-RH) (Fig. 9).
Because our initial experiments showed that ventral hemisections caused qualitatively larger impairments in interlimb coordination and hindlimb (hip) cycle period than dorsal hemisections (Fig. 3, C and D), we categorized experiments based on damage to the VLF when analyzing these parameters. Experiments in which the VLF was bilaterally spared (n = 8; all experiments in Fig. 2, A, C, and H) had the smallest effects on the strength of interlimb phase coupling. Postlesion swimming displayed significant Rao P values and mean phases similar to prelesion (Figs. 7B and 8A). Mean vector lengths also indicated strong coupling (Fig. 9A). All of the turtles in the postlesion "bilaterally spared VLF" group exhibited tightly clustered, statistically significant distributions in all four sets of phase measurements (LF-RF, RH-RF, LH-RF, LH-RH) (P < 0.05, Rao spacing test), despite small but significant shifts from prelesion means (Watson–Williams test, P < 0.05, Table 1). Pooled mean vector lengths for each phase measurement were 0.88 ± 0.06, 0.85 ± 0.16, 0.88 ± 0.07, and 0.81 ± 0.19, respectively (Fig. 9A).
Lesions that bilaterally damaged the VLF (n = 16) generally caused impairments in phase coupling between the right forelimb and both hips (Figs. 3D and 4D). Six of these turtles, each with especially extensive lesions that included VLF and DLF on both sides, did not perform any swim movements in one or both hips postlesion (see Movement amplitudes) and were excluded from further analyses. The remaining animals continued to exhibit postlesion swim movements in both hindlimbs, although these movements were often poorly coordinated with the forelimbs (n = 10; experiments R38, R44, R48, R51–R52, R60–R61, R73–R74, and R78 in Fig. 2, B, E, and G). Only one of these ten preparations (experiment R52, Table 2), in which the lesion left some right and left VLF intact (Fig. 2B), displayed significant forelimb–hindlimb coordination on both sides after surgery. The other nine turtles lost significant forelimb–hindlimb coordination either bilaterally (n = 5; Table 2), or unilaterally (n = 4). The strength of postlesion forelimb–hindlimb coupling was greatly reduced on both sides in this group compared with prelesion controls, as indicated by pooled mean vector lengths for RH-RF and LH-RF phases that changed from 0.95 ± 0.03 and 0.94 ± 0.03 prelesion to 0.51 ± 0.25 and 0.50 ± 0.24 postlesion, respectively (Fig. 9A). However, it was noteworthy that right–left hip coordination, although weakened relative to prelesion controls (pooled mean vector lengths of 0.93 ± 0.03 and 0.60 ± 0.25 pre- and postlesion, respectively; Fig. 9A), remained significant in six of ten experiments within the group (Table 2). The most striking example of this unusual condition, in which forelimb–hindlimb coupling was lost or greatly reduced, whereas right–left coordination remained, was experiment R38 (Fig. 7C), in which fast-alternating forelimbs and slow-alternating hindlimbs swam at their own speeds during locomotor episodes and were uncoordinated with each other.
|
CYCLE PERIOD ANALYSIS.
We analyzed cycle period as an additional measure of interlimb coordination. Swim episodes were analyzed only if they exhibited forelimb cycle periods 
1.5 s. Prior to spinal lesions, all limbs exhibited 1:1 coordination with each other, with approximately equal cycle periods during any given episode (Figs. 3A and 4A).
Like our phase data, cycle period values were grouped by experiment according to whether the spinal lesion included bilateral or unilateral VLF damage. Lesions that bilaterally spared the VLF (n = 8; see experiments listed earlier) had no significant effect on the mean hip cycle period within that group (Fig. 10A). In contrast, lesions that caused bilateral VLF damage (n = 10) increased the mean cycle period of both hips relative to the right forelimb (Fig. 10A). In two of these experiments, where the VLF was bilaterally damaged but other regions of the cord were largely spared (experiments R74 and R78 in Fig. 2, E and G), we observed a bilateral increase in hip cycle period relative to forelimb (R74: RH-RF ratio = 1.81 ± 0.21, LH-RF ratio = 2.04 ± 0.20; R78: RH-RF ratio = 2.15 ± 0.22, LH-RF ratio = 1.96 ± 0.29). Finally, we found that unilateral VLF lesions (n = 16) had greater effects on the cycle period of the ipsilateral hip than of the contralateral hip (Fig. 10B). The data in Fig. 10B show that the hip contralateral to the lesion tended to retain the same cycle period as the forelimbs (hip/forelimb period ratios 
1.0), whereas the ipsilateral hip tended to slow down (display a lengthened cycle period) relative to the forelimbs (hip/forelimb period ratios >1.0).
|
|
DISCUSSION |
|---|
|
; Lennard and Stein 1977) or contralateral fictive forward swim motor patterns in hindlimb nerves (Juranek and Currie 2000), by artificially activating the same locomotor command pathways that the animal uses to initiate and control voluntary locomotion. We also showed that interenlargement pathways in the VLF have a critical role in coordinating swimming movements between ipsilateral limbs (forelimb–hindlimb) as well as between the right and left hindlimbs. Spinal lesions that bilaterally spared the VLF had little to no effect on interlimb coupling or cycle period (Table 1, Figs. 7B, 8, A and C, 9A, and 10A). Lesions that bilaterally damaged the VLF weakened forelimb–hindlimb coupling on both sides (Table 2, Figs. 7C, 8B, 9A, and 10A), but frequently had less pronounced effects on right–left hindlimb coordination, so long as roughly equal amounts of ventral white matter were severed on the left and right sides. Finally, lesions that damaged the VLF unilaterally resulted in a loss of phase coupling between the forelimbs and the hindlimb ipsilateral to the lesion as well as between right and left hindlimbs, with little effect on coupling between the forelimbs and the hindlimb contralateral to the lesion (Table 3, Figs. 7A, 9B, and 10B). There is no evidence that the brief 90° twisting of the D2–D3 cord that was required to make dorsal and ventral hemisections itself produced any cord damage. The D2 and D3 spinal roots were cut before twisting and lesioning, so that there was considerable slack in the D2–D3 region (see Lesions under METHODS). Also, given that the cord was twisted in the same direction (toward the animal's left) for dorsal and ventral hemisections, any twisting-induced deficits should have been the same between the dorsal and ventral hemisection groups. The strikingly different effects produced by these lesions support the conclusion that the cord twisting itself did not cause damage. Our findings are summarized in Fig. 11. Together with our previous paper (Samara and Currie 2007), these experiments show the importance of examining in situ neural networks during voluntary animal behaviors along with cellular and synaptic studies in artificially stimulated in vitro preparations.
|
It has been suggested, based on electrical stimulation studies in intact and spinalized turtles, that reticulospinal pathways traveling in the DLF carry descending locomotor commands to the contralateral hindlimb (Juranek and Currie 2000; Lennard and Stein 1977) and forelimb (Stein 1978). That work demonstrated that artificial activation of DLF fibers was sufficient to evoke a facsimile of forward swimming in the contralateral limb(s). The reticular origin of these crossed commands is supported by the fact that electrical stimulation in the lateral reticular formation (RF) of the turtle brain stem also evoked swimming in the contralateral forelimb and hindlimb of animals with movement (Kazennikov et al. 1980) and fictive swimming in the contralateral forelimb and hindlimb nerves of immobilized preparations (Currie 2003). The locations of these reticular "locomotor points" partially overlapped with the positions of magnocellular RF cell bodies that were observed in retrograde tracing studies, in which damage to the lateral white matter of the cervical cord labeled somata throughout the reticular formation and several other supraspinal regions (ten Donkelaar 1976a; ten Donkelaar et al. 1980). Using anterograde degeneration, ten Donkelaar (1976b) demonstrated that some reticulospinal axons in the turtle spinal cord descended through the lateral funiculus in an area corresponding to that which produced swim movements in response to electrical stimulation (compare Fig. 1 in Lennard and Stein 1977 and Fig. 8 in ten Donkelaar 1976b).
In addition to reticulospinal axons, rubrospinal axons in reptiles also occur in the DLF, although in a more dorsal position, occupying lateral white matter near the upper half of the dorsal horn, whereas reticulospinal fibers span a region of lateral white matter from the lower half of the dorsal horn through the upper half of the ventral horn (ten Donkelaar 1976b). Rubrospinal axons have been shown to project from the contralateral red nucleus as far as the lumbar spinal cord in turtles (Woodson and Künzle 1982) and are thought to underlie rapid locomotor responses to visual stimuli in lizards (Martínez-Marcos et al. 1999). It is possible then that the destruction of rubrospinal axons may have contributed to impaired hip movement amplitudes following DLF lesions in our study.
Our data indicate that turtle locomotor pathways that activate voluntary forward swimming in the hindlimbs are concentrated in the DLF at the midbody (D2–D3) level, but also present in smaller numbers within the adjacent VLF. This is based on the following findings: 1) bilateral destruction of the DLF but sparing of the VLF greatly reduced the amplitude but did not abolish hindlimb swimming movements (Figs. 3C and 4C), indicating that some command function remained in the spared ventral white matter; 2) bilateral destruction of the VLF but sparing of the DLF had no significant effect on hindlimb swim amplitudes (Figs. 3D and 4D), suggesting that DLF axons by themselves were sufficient; and 3) lesions that destroyed part or all of the DLF and VLF on both sides completely abolished hindlimb swimming (Fig. 2G, experiments R72, R40–R42, R66, and R69), whereas lesions that spared only the DLF and VLF on both sides (Figs. 2A and 4B) permitted normal swimming in both hindlimbs. Thus electrical stimulation of sites in the midbody DLF might activate hindlimb swimming movements whereas stimulation in VLF does not (Juranek and Currie 2000; Lennard and Stein 1977) because of a higher density of locomotion-activating axons in the DLF. An alternative explanation for these same findings is that pathways activating locomotion are totally confined to the DLF, but rhythmically active propriospinal axons that descend from the forelimb enlargement in the VLF, and underlie forelimb–hindlimb coordination, are also able to drive small alternating hindlimb movements in the absence of brain stem commands.
Our results differ from mammalian spinal lesion studies, which showed that tracts in the ventrolateral white matter were necessary for voluntary control of the hindlimbs during locomotion (Eidelberg 1981; Jordan 1986; Orlovsky et al. 1999; Steeves and Jordan 1980). Lesion studies from a number of species suggest parallel locomotor pathways that travel through both the ventrolateral and dorsolateral white matter. For example, lesioning the VLF or DLF alone permitted lamprey swimming, whereas cutting all the lateral white matter did not (McClellan 1988). Studies in rats (Loy et al. 2002b) and cats (Yamaguchi 1986) suggested that VLF lesions alone had much less of an effect than combined VLF–DLF lesions. Other lesion studies in cats showed the involvement of DLF pathways in PLR-evoked locomotion (Noga et al. 1991) and VLF pathways in MLR-evoked (Steeves and Jordan 1980) and PLR-evoked (Noga et al. 1991) locomotion. Additional evidence for diffusely spread locomotor pathways in lateral white matter comes from spinal cord electrical stimulation work, as stimulation of the DLF (cat: Kazennikov et al. 1983; Sherrington 1910; Yamaguchi 1986; stingray: Williams et al. 1984) as well as the VLF (cat: Yamaguchi 1986; stingray: Williams et al. 1984) elicited locomotor movements.
Our data support the conclusion that turtle DLF command pathways activate voluntary swimming primarily in the contralateral hindlimb, but also weakly activate ipsilateral swimming movements. This is partly based on the observation that right lateral hemisection of the D2–D3 cord greatly reduced the amplitude but did not completely abolish left (contralateral) hindlimb swimming movements (Fig. 3B), indicating that some uncrossed locomotor signal in the spared left hemicord was able to drive weak left-side swimming movements. However, note that when only a small portion of the right lateral white matter (mainly DLF) was spared, the turtle exhibited only slow left-side (contralateral to the spared white matter) hindlimb swimming, with no right-side hindlimb movements at all (Fig. 6). Currently it is not known whether different reticulospinal cell populations mediate these crossed and uncrossed locomotor commands. In turtles, the finding that mainly contralateral swimming movements are driven by descending locomotor commands is similar to the crossed activation of contralateral swimming in stingray pectoral fins (Livingston and Leonard 1990; Williams et al. 1984). However, in most other vertebrates that have been examined, descending reticulospinal signals activated locomotor movements bilaterally (lampreys: McClellan 1988; Wannier et al. 1998), with larger ipsilateral movements predominating in limbed species (geese: Sholomenko and Steeves 1987; cat: Noga et al. 1991; Sherrington 1910; Steeves and Jordan 1980).
We currently do not know precisely where turtle DLF locomotor commands cross the midline. Preliminary evidence suggests that hindlimb locomotor commands descend ipsilateral to the D3 DLF stimulation site at least as far as segment D5 (Currie 2000). It is likely that a large fraction of the command signal crosses anterior to the hindlimb enlargement (segments D8–S2), since D7–S2 bisections (midsagittal lesions that split the cord along the longitudinal midline) reduced hindlimb swim amplitudes more severely than D8–S2 bisections in voluntarily swimming turtles (Samara and Currie 2007). Also, stimulation of the right DLF continued to evoke swimlike movements and EMG activity in the left hindlimb after surgical removal of the right D8–S2 hemicord (Samara and Currie 2006). Given these results, it is likely that DLF command signals cross mainly in segment D7 and/or segment D6. However, further experiments are needed to verify this and to identify and characterize the presumed reticulospinal neurons that activate and maintain turtle locomotion. Some limited progress has already been made in characterizing the interactions between DLF commands and presumed hindlimb pattern generation circuitry. Berkowitz (2002) recorded extracellularly from ventral horn interneurons in segment D8 of the anterior turtle hindlimb enlargement and found that the majority of cells were active during fictive swimming evoked by electrical stimulation of the contralateral D3 DLF and during fictive scratching evoked by cutaneous stimulation; this result indicated that the DLF swim command activates D8 interneurons that are shared by swim and scratch motor pattern generating networks. Anterior segments of the hindlimb enlargement are known to be most important for hindlimb scratch rhythmogenesis in turtles (Mortin and Stein 1989) and to receive greater descending input than more posterior segments (Berkowitz 2004).
Pathways contributing to interlimb coordination
Previous studies showed that high-spinal turtles with movement typically displayed 1:1 forelimb–hindlimb coordination during swimlike movements evoked by repetitive DLF stimulation (Stein 1978), suggesting that sufficient circuitry for forelimb–hindlimb coordination resided within the spinal cord. In retrograde tracing studies, Kusuma and ten Donkelaar (1980) found that descending and ascending propriospinal fibers in the turtle midbody spinal cord traveled through the ventral and lateral funiculi, the vast majority being ipsilateral to the lesion site. The fact that these fibers connected the limb enlargements (Kusuma and ten Donkelaar 1980) implicated them in maintaining proper forelimb–hindlimb phase coupling. Our results are consistent with these populations mediating forelimb–hindlimb coordination, since lesions that unilaterally damaged the midbody VLF most severely affected coordination between the forelimbs and the hip ipsilateral to the lesion (Table 3, Figs. 9B and 10B).
Although our results implicate axons of the midbody VLF in forelimb–hindlimb coupling, other areas of white matter might also contain coordinating units. In cats, presumed propriospinal fibers that contributed to forelimb–hindlimb coordination appeared to be broadly distributed in the lateral white matter, since only lesions that bilaterally damaged both the VLF and the DLF completely decoupled forelimb and hindlimb locomotor rhythms (Bem et al. 1996). Rats also displayed a diffusely distributed coordinating tract within the VLF and DLF because damage to either region impaired forelimb–hindlimb coordination during grid walking (Schucht et al. 2002) and open field walking (Loy et al. 2002b). Similar to our results, thoracic hemisection experiments suggested that these pathways serve to coordinate the forelimbs with the hindlimb ipsilateral to the lesion (Kato 1992). However, unlike the data from cats, our experiments showed a loss of forelimb–hindlimb coupling after VLF lesions, even if the DLF was bilaterally spared (Table 2, experiments R38 and R48). In contrast, we did not observe impairment in forelimb–hindlimb coordination when the DLF but not the VLF was damaged (Table 1, experiments R45–R64; Figs. 3C, 4C, 7B, and 8A). Our results show that the VLF plays a significant role in forelimb–hindlimb coordination during turtle swimming, whereas the DLF does not. Our results also show that tracts within the ventral funiculus (VF), ventral and/or medial to the ventral horn, are not necessary for forelimb–hindlimb coordination. The VF has been shown to contain propriospinal fibers that connect the limb enlargements in turtles (Kusuma and ten Donkelaar 1980). However, lesions such as R76 and R77 (Figs. 2A and 4B), which damaged the VF and other medial white matter, but left the VLF intact, did not impair interlimb phase coupling or hip cycle period. Thus our results indicate that axon tracts mediating ipsilateral forelimb–hindlimb coordination during turtle swimming are concentrated in the VLF of the interenlargement cord.
In addition to forelimb–hindlimb coordination, our results also suggest a role for longitudinal propriospinal VLF pathways in maintaining proper out-of-phase coordination between the right and left hindlimbs. Previous data from our lab (Samara and Currie 2007) demonstrated that longitudinal spinal pathways were sufficient to maintain right–left hindlimb alternation after all commissural connections in the hindlimb enlargement and first preenlargement segment (segments D7–S2) were severed by a midsagittal lesion that completely separated the right and left halves of the posterior cord. In the current study, spinal lesions that unilaterally damaged the VLF disrupted not only forelimb–hindlimb coupling ipsilateral to the lesion, but also right–left hindlimb coordination (Figs. 3B, 7A, 9B, and 10B). The hips lost significant right–left coupling in 10 of 16 experiments with unilateral VLF damage (Table 3). Even bilateral VLF lesions with different dorsoventral extents between the right and left cords (experiments R61, R73, and R78, Fig. 2, E and G) caused a loss of significant LH-RH coupling (Table 2). This implies that an imbalance between forelimb–hindlimb coupling on the right and left sides is an important factor in the loss of right–left hip coordination and, conversely, suggests that intact forelimb–hindlimb coupling helps to maintain normal out-of-phase right–left hip coordination. In addition, our results imply that dorsal pathways play little if any role in hindlimb–hindlimb coordination. Lesions that bilaterally spared the VLF, even with different dorsoventral extents between the right and left sides, always permitted significant LH-RH coupling (Table 1). Data from other vertebrates also support the hypothesis that descending pathways can contribute to right–left hindlimb coordination. With descending circuitry intact, surgically splitting the lumbar enlargement did not eliminate 1:1 right–left alternation between lumbar ventral roots in isolated neonatal rat spinal cords (Cowley and Schmidt 1997; Kremer and Lev-Tov 1997) or between the hindlimbs in voluntarily walking cats (Kato 1988). These descending pathways are likely to be propriospinal because spinalization did not affect interlimb coordination during treadmill locomotion (cat: Miller and van der Meche 1976; chick: Bekoff et al. 1989) and fictive locomotion (chick: Jacobson and Hollyday 1982).
Although our current results and previous lesion study (Samara and Currie 2007) both suggested that interenlargement propriospinal pathways in the VLF contribute to hindlimb–hindlimb coordination, we do not believe that these longitudinal pathways act alone in maintaining right–left hindlimb alternation. Extensive evidence indicates that crossed commissural fibers in the turtle hindlimb enlargement can also contribute to right–left coordination during scratch reflex (Currie and Gonsalves 1997, 1999; Currie and Lee 1997; Currie and Stein 1989; Stein et al. 1995, 1998), flexion reflex (Currie and Lee 1996), and chemically activated hindlimb motor patterns in isolated turtle spinal cords (Currie 1999). New experiments are needed to gauge the relative contributions of longitudinal (interenlargement) propriospinal and crossed commissural coordinating mechanisms during turtle locomotion and to identify and characterize the spinal interneurons that underlie these functions.
|
|
GRANTS |
|---|
|
|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: S. N. Currie, Department of Cell Biology and Neuroscience, University of California–Riverside, Riverside, CA 92521 (E-mail: currie{at}mail.ucr.edu)
|
|
REFERENCES |
|---|
|
Bekoff A, Kauer JA, Fulstone A, Summers TR. Neural control of limb coordination. II. Hatching and walking motor output patterns in the absence of input from the brain. Exp Brain Res 74: 609–617, 1989.[Web of Science][Medline]
Bem T, Gorska T, Majczynski H, 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.[Web of Science][Medline]
Berkowitz A. Both shared and specialized spinal circuitry for scratching and swimming in turtles. J Comp Physiol A Sens Neural Behav Physiol 188: 225–234, 2002.[CrossRef][Web of Science][Medline]
Berkowitz A. Propriospinal projections to the ventral horn of the rostral and caudal hindlimb enlargement in turtles. Brain Res 1014: 164–176, 2004.[CrossRef][Web of Science][Medline]
Berkowitz A, Stein PSG. Activity of descending propriospinal axons in the turtle hindlimb enlargement during two forms of fictive scratching: phase analyses. J Neurosci 14: 5105–5119, 1994.[Abstract]
Bernau NA, Puzdrowski RL, Leonard RB. Identification of the midbrain locomotor region and its relation to descending locomotor pathways in the Atlantic stingray, Dasyatis sabina. Brain Res 557: 83–94, 1991.[CrossRef][Web of Science][Medline]
Brustein E, 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.
Buchanan JT. Commissural interneurons in rhythm generation and intersegmental coupling in the lamprey spinal cord. J Neurophysiol 81: 2037–2045, 1999.
Cowley KC, Schmidt BJ. Effects of inhibitory amino acid antagonists on reciprocal inhibitory interactions during rhythmic motor activity in the in vitro neonatal rat spinal cord. J Neurophysiol 74: 1109–1117, 1995.
Cowley KC, Schmidt BJ. Regional distribution of the locomotor pattern-generating network in the neonatal rat spinal cord. J Neurophysiol 77: 247–259, 1997.
Currie SN. Fictive hindlimb motor patterns evoked by AMPA and NMDA in turtle spinal cord-hindlimb nerve preparations. J Physiol (Paris) 93: 199–211, 1999.
Currie SN. Descending pathways that activate fictive swim motor patterns in the turtle spinal cord. Soc Neurosci Abstr 26: 746.1, 2000.
Currie SN. Fictive locomotion evoked by electrical stimulation of the brainstem in decerebrate immobilized turtles. Soc Neurosci Abstr 29: 276.9, 2003.
Currie SN, Gonsalves GG. Right–left interactions between rostral scratch networks generate rhythmicity in the preenlargement spinal cord of the turtle. J Neurophysiol 78: 3479–3483, 1997.
Currie SN, Gonsalves GG. Reciprocal interactions in the turtle hindlimb enlargement contribute to scratch rhythmogenesis. J Neurophysiol 81: 2977–2987, 1999.
Currie SN, Lee S. Glycinergic inhibition in the turtle spinal cord regulates the intensity and pattern of fictive flexion reflex motor output. Neurosci Lett 205: 75–78, 1996.[CrossRef][Web of Science][Medline]
Currie SN, Lee S. Glycinergic inhibition contributes to the generation of rostral scratch motor patterns in the turtle spinal cord. J Neurosci 17: 3322–3333, 1997.
Currie SN, Stein PSG. Interruptions of fictive scratch motor rhythms by activation of cutaneous flexion reflex afferents in the turtle. J Neurosci 9: 488–496, 1989.[Abstract]
Earhart GM, Stein PSG. Step, swim, and scratch motor patterns in the turtle. J Neurophysiol 84: 2181–2190, 2000.
Eidelberg E. Consequences of spinal cord lesions upon motor function, with special reference to locomotor activity. Prog Neurobiol 17: 185–202, 1981.[CrossRef][Web of Science][Medline]
Eidelberg E, Walden JG, Nguyen LH. Locomotor control in macaque monkeys. Brain 104: 647–663, 1981.
Field EC, Stein PSG. Spinal cord coordination of hindlimb movements in the turtle: interlimb temporal relationships during bilateral scratching and swimming. J Neurophysiol 78: 1404–1413, 1997.
Giovanelli Barilari M, Kuypers HG. Propriospinal fibers interconnecting the spinal enlargements in the cat. Brain Res 14: 321–330, 1969.[CrossRef][Web of Science][Medline]
Gorska T, Bem T, Majczynski H, Zmyslowski W. Unrestrained walking in cats with partial spinal lesions. Brain Res Bull 32: 241–249, 1993.[CrossRef][Web of Science][Medline]
Gorska T, Bem T, Majczynski H, Zmyslowski W. Different forms of impairment of the fore–hindlimb coordination after partial spinal lesions in cats. Acta Neurobiol Exp 56: 177–188, 1996.[Medline]
Gorska T, Majczynski H, Bem T, Zmyslowski W. Hindlimb swing, stance and step relationships during unrestrained walking in cats with lateral funicular lesion. Acta Neurobiol Exp 53: 133–142, 1993.[Medline]
Grillner S, Zangger P. On the central generation of locomotion in the low spinal cat. Exp Brain Res 34: 241–261, 1979.[Web of Science][Medline]
Jacobson RD, Hollyday M. Electrically evoked walking and fictive locomotion in the chick. J Neurophysiol 48: 257–270, 1982.
Jiang W, 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 of locomotion in the adult cat. I. Treadmill walking. J Neurophysiol 76: 849–866, 1996.
Jordan LM. Initiation of locomotion from the mammalian brainstem. In: Neurobiology of Vertebrate Locomotion (Wenner-Gren Centre International Symposium Series, vol. 45), edited by Grillner S, Stein PSG, Stuart DG, Forssberg H, Herman RM. London: Macmillan, 1986, p. 21–37.
Juranek J, Currie SN. Electrically evoked fictive swimming in the low-spinal immobilized turtle. J Neurophysiol 83: 146–155, 2000.
Juvin L, Simmers J, Morin D. Propriospinal circuitry underlying interlimb coordination in mammalian quadrupedal locomotion. J Neurosci 25: 6025–6035, 2005.
Kato M. Longitudinal myelotomy of lumbar spinal cord has little effect on coordinated locomotor activities of bilateral hindlimbs of the chronic cats. Neurosci Lett 93: 259–263, 1988.[CrossRef][Web of Science][Medline]
Kato M. Walking of cats on a grid: performance of locomotor task in spinal intact and hemisected cats. Neurosci Lett 145: 129–132, 1992.[CrossRef][Web of Science][Medline]
Kazennikov OV, Selionov VA, Shik ML, Iakovleva GV. Rhombencephalic "locomotor area" of turtles. Neirofiziologiia 12: 382–390, 1980.[Medline]
Kazennikov OV, Shik ML, Iakovleva GV. Stepping movements induced in cats by stimulation of the dorsolateral funiculus of the spinal cord. Bull Exp Biol Med 96: 1036–1039, 1983.[CrossRef][Web of Science]
Kremer E, Lev-Tov A. Localization of the spinal network associated with generation of hindlimb locomotion in the neonatal rat and organization of its transverse coupling system. J Neurophysiol 77: 1155–1170, 1997.
Kusuma A, ten Donkelaar HJ. Propriospinal fibers interconnecting the spinal enlargements in some quadrupedal reptiles. J Comp Neurol 193: 871–891, 1980.[CrossRef][Web of Science][Medline]
Lennard PR. Afferent perturbations during "monopodal" swimming movements in the turtle: phase-dependent cutaneous modulation and proprioceptive resetting of the locomotor rhythm. J Neurosci 5: 1434–1435, 1985.[Abstract]
Lennard PR, Stein PS. Swimming movements elicited by electrical stimulation of turtle spinal cord. I. Low-spinal and intact preparations. J Neurophysiol 40: 768–778, 1977.
Levitin DJ, Russell GS. Rao's spacing test. In: Encyclopedia of Statistical Sciences Update, edited by Kotz S, Read CB, Banks DL. New York: Wiley, 1999, vol. 3, p. 87–89.
Livingston CA. The Descending Pathways That Control Locomotion in the Atlantic Stingray, Dasyatis sabina (PhD dissertation). Galveston, TX: Univ. of Texas Medical Branch at Galveston, 1987.
Livingston CA, Leonard RB. Locomotion evoked by stimulation of the brain stem in the Atlantic stingray, Dasyatis sabina. J Neurosci 10: 194–204, 1990.[Abstract]
Loy DN, Magnuson DS, Zhang YP, Onifer SM, Mills MD, Cao QL, Darnall JB, Fajardo LC, Burke DA, Whittemore SR. Functional redundancy of ventral spinal locomotor pathways. J Neurosci 22: 315–323, 2002a.
Loy DN, Talbott JF, Onifer SM, Mills MD, Burke DA, Dennison JB, Fajardo LC, Magnuson DS, Whittemore SR. Both dorsal and ventral spinal cord pathways contribute to overground locomotion in the adult rat. Exp Neurol 177: 575–580, 2002b.[CrossRef][Web of Science][Medline]
Magnuson DS, Trinder TC. Locomotor rhythm evoked by ventrolateral funiculus stimulation in the neonatal rat spinal cord in vitro. J Neurophysiol 77: 200–206, 1997.
Mardia KV, Jupp PE. Directional Statistics (2nd ed.). New York: Wiley, 2000.
Martínez-Marcos A, Lanuza E, Font C, Martínez-García F. Afferents to the red nucleus in the lizard Podarcis hispanica: putative pathways for visuomotor integration. J Comp Neurol 411: 35–55, 1999.[CrossRef][Web of Science][Medline]
McClellan AD. Brainstem command systems for locomotion in the lamprey: localization of descending pathways in the spinal cord. Brain Res 457: 338–349, 1988.[CrossRef][Web of Science][Medline]
Miller S, van der Meche FGA. Coordinated stepping of all four limbs in the high spinal cat. Brain Res 109: 395–398, 1976.[CrossRef][Web of Science][Medline]
Mortin LI, Stein PSG. Spinal cord segments containing key elements of the central pattern generators for three forms of scratch reflex in the turtle. J Neurosci 9: 2285–2296, 1989.[Abstract]
Noga BR, Kriellaars DJ, Jordan LM. The effect of selective brainstem or spinal cord lesions on treadmill locomotion evoked by stimulation of the mesencephalic or pontomedullary locomotor regions. J Neurosci 11: 1691–1700, 1991.[Abstract]
Nyberg-Hansen R. Functional organization of descending supraspinal fibre systems to the spinal cord. Anatomical observations and physiological correlations. Ergeb Anat Entwicklungsgesch 39: 3–48, 1966.[Medline]
Orlovsky GN, Deliagina TG, Grillner S. Initiation of locomotion. In: Neuronal Control of Locomotion: From Mollusc to Man. New York: Oxford Univ. Press, 1999, p. 205–214.
Reed WR, Shum-Siu A, Onifer SM, Magnuson DS. Inter-enlargement pathways in the ventrolateral funiculus of the adult rat spinal cord. Neuroscience 142: 1195–1207, 2006.[CrossRef][Web of Science][Medline]
Samara RF, Currie SN. Impairment of voluntary locomotion in turtles following partial lesions of the mid-body spinal cord. Soc Neurosci Abstr 30: 882.12, 2004.
Samara RF, Currie SN. Electrical activation of hindlimb locomotor activity in the turtle spinal cord hemi-enlargement preparation. Soc Neurosci Abstr 32: 448.27, 2006.
Samara RF, Currie SN. Crossed commissural pathways in the spinal hindlimb enlargement are not necessary for right–left hindlimb alternation during turtle swimming. J Neurophysiol 98: 2223–2231, 2007.
Schucht P, Raineteau O, Schwab ME, Fouad K. Anatomical correlates of locomotor recovery following dorsal and ventral lesions of the rat spinal cord. Exp Neurol 176: 143–153, 2002.[CrossRef][Web of Science][Medline]
Sherrington CS. Flexion-reflex of the limb, crossed extension-reflex, and reflex stepping and standing. J Physiol 40: 28–121, 1910.
Sholomenko GN, Steeves JD. Effects of selective spinal cord lesions on hind limb locomotion in birds. Exp Neurol 95: 403–418, 1987.[CrossRef][Web of Science][Medline]
Siegel S. Nonparametric Statistics for the Behavioral Sciences. New York: McGraw-Hill, 1956.
Steeves JD, Jordan LM. Localization of a descending pathway in the spinal cord which is necessary for controlled treadmill locomotion. Neurosci Lett 20: 283–288, 1980.[CrossRef][Web of Science][Medline]
Steeves JD, Sholomenko GN, Webster DM. Stimulation of the pontomedullary reticular formation initiates locomotion in decerebrate birds. Brain Res 401: 205–212, 1987.[CrossRef][Web of Science][Medline]
Stein PSG. Swimming movements elicited by electrical stimulation of the turtle spinal cord: the high spinal preparation. J Comp Physiol A Sens Neural Behav Physiol 124: 203–210, 1978.[CrossRef]
Stein PSG, McCullough ML, Currie SN. Reconstruction of flexor/extensor alternation during fictive rostral scratching by two-site stimulation in the spinal turtle with a transverse spinal hemisection. J Neurosci 18: 467–479, 1998.
Stein PSG, Victor JC, Field EC, Currie SN. Bilateral control of hindlimb scratching in the spinal turtle: contralateral circuitry contributes to the normal ipsilateral motor pattern of fictive rostral scratching. J Neurosci 15: 4343–4355, 1995.[Abstract]
Tegner J, Matsushima T, el Manira A, Grillner S. The spinal GABA system modulates burst frequency and intersegmental coordination in the lamprey: differential effects of GABAA and GABAB receptors. J Neurophysiol 69: 647–657, 1993.
ten Donkelaar HJ. Descending pathways from the brain stem to the spinal cord in some reptiles. I. Origin. J Comp Neurol 167: 421–442, 1976a.[CrossRef][Web of Science][Medline]
ten Donkelaar HJ. Descending pathways from the brain stem to the spinal cord in some reptiles. II. Course and site of termination. J Comp Neurol 167: 443–464, 1976b.[CrossRef][Web of Science][Medline]
ten Donkelaar HJ, Kusuma A, de Boer-Van Huizen R. Cells of origin of pathways descending to the spinal cord in some quadrupedal reptiles. J Comp Neurol 192: 827–851, 1980.[CrossRef][Web of Science][Medline]
von Holst E. Relative coordination as a phenomenon and as a method of analysis of central nervous system function (translated from German). In: The Behavioural Physiology of Animals and Man, edited by Martin RD. Coral Gables, FL: Univ. of Miami Press, et al. 1939/1973, vol. 1, p. 139–173.
Walker WF. A structural and functional analysis of walking in the turtle, Chrysemys picta marginata. J Morph 134: 195–214, 1971.[CrossRef][Medline]
Wannier T, Deliagina TG, Orlovsky GN, Grillner S. Differential effects of the reticulospinal system on locomotion in lamprey. J Neurophysiol 80: 103–112, 1998.
Webster DM, Steeves JD. Funicular organization of avian brainstem-spinal projections. J Comp Neurol 312: 467–476, 1991.[CrossRef][Web of Science][Medline]
Williams BJ, Livingston CA, Leonard RB. Spinal cord pathways involved in initiation of swimming in the stingray, Dasyatis sabina: spinal cord stimulation and lesions. J Neurophysiol 51: 578–591, 1984.
Woodson W, Künzel H. Distribution and structural characterization of neurons giving rise to descending spinal projections in the turtle, Pseudemys scripta elegans. J Comp Neurol 212: 336–348, 1982.[CrossRef][Web of Science][Medline]
Yamaguchi T. Descending pathways eliciting forelimb stepping in the lateral funiculus: experimental studies with stimulation and lesion of the cervical cord in decerebrate cats. Brain Res 379: 125–136, 1986.[CrossRef][Web of Science][Medline]
Zar JH. Biostatistical Analysis (4th ed.). Upper Saddle River, NJ: Prentice Hall, 1999.
Zmyslowski W, Gorska T, Majczynski H, Bem T. Hindlimb muscle activity during unrestrained walking in cats with lesions of the lateral funiculi. Acta Neurobiol Exp 53: 143–153, 1993.[Medline]
Zug GR. Buoyancy, locomotion, morphology of the pelvic girdle and hindlimbs, and systematics of cryptodiran turtles. Misc Publ Mus Zool Univ Mich 142: 1–98, 1971.
This article has been cited by other articles:
|
|
 |
|
  D. S. Marigold and J. E. Misiaszek Whole-Body Responses: Neural Control and Implications for Rehabilitation and Fall Prevention Neuroscientist, February 1, 2009; 15(1): 36 - 46. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
TOP
ABSTRACT
INTRODUCTION
