- Email: [email protected]
Effects of infant versus adult pyramidal tract lesions on locomotor behavior in hamsters
EXPERIMENTAL
NEUROLOGY
111,98-105
(19%)
Effects of Infant versus Adult Pyramidal Tract Lesions on Locomotor Behavior in Hamsters JOYCE KEIFER’ Neuroscience
Training
Program
and Department
AND KATHERINE of Anatomy,
MATERIALS
INTRODUCTION
0014.4886/91
$3.00
Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
Madison,
Wzkconsin
53706
AND
METHODS
Fifteen normal adult golden hamsters (Mesocricetus aurutus) ranging in age from 3 weeks to 6 months were filmed as they performed locomotion. Anatomically, the pyramidal tract of hamsters is well developed by 3 weeks of age (42) and hamsters show a mature form of
A major descending pathway, the pyramidal tract, arises from the sensorimotor cortex and in most mamof Physiology, Chicago Avenue,
of Wisconsin,
mals projects to all segments of the spinal cord. Previous studies of the pyramidal tract have demonstrated its importance in behaviors that require a high degree of motor skill, particularly during fine manipulatory movements of the forepaw and digits. For example, after a lesion of the medullary pyramids, monkeys are unable to use the digits independently (34). Rodents with a pyramidal tract lesion exhibit deficits in placing (13,25) and fine control of the forepaw (25,27,43). Cats with pyramidal tract lesions are also impaired in forelimb reaching movements (21). However, because of its extensive projections to the dorsal horn of the spinal cord, the pyramidal tract may also play a more general role in modulating behaviors requiring sensory feedback. Thus, even those behaviors that are relatively stereotyped, such as locomotion, may be influenced by descending corticospinal input. Several studies suggest that the pyramidal tract does play a role in locomotion. Observations of animals with pyramidal tract lesions reveal disorders in posture and gait, especially when skill beyond the mere rhythm of locomotion is required (2, 8, 14, 25, 36). The aims of the present study were twofold. First, given the evidence that the pyramidal tract is required for skilled movements involving sensory feedback and that it may also play a role in locomotion, we wished to determine how lesions of the pyramidal tract might disrupt joint movements and foot placements during locomotion in a situation requiring continual adjustment to a changing terrain. The second goal was to determine whether animals with lesions as infants, in which anatomical rearrangements are known to occur after pyramidal tract lesions (3, 4, 28, 32), would show greater functional recovery in locomotor behavior than animals with adult pyramidotomy. Some of the results have been previously reported (30, 31).
The role of the pyramidal tract in locomotion was studied in hamsters by analyzing their locomotor behavior after lesions of the medullary pyramidal tract. Animals with lesions either as adults or as infants were compared to determine whether early pyramidotomy results in greater functional recovery. Normal and pyramidotomized animals were filmed during locomotion on a runway consisting of either smooth or rough terrain to assess whether the uneven surface would accentuate locomotor deficits. Frame-by-frame analysis of the filmed behavior during all phases of the step cycle was carried out to determine positions of the joints of the forelimb and hindlimb during locomotion. Accuracy of limb placement on the rough terrain was determined by observations of consecutive step cycles. The results show that lesions of the pyramidal tract in both infant and adult hamsters affect locomotion first by causing a reduction in the yielding phase of the step cycle and second by producing inaccuracies of forelimb placement. Rough terrain accentuates deficits in forelimb placement during locomotion. Animals with lesions as infants and those with lesions as adults show surprisingly similar deficits in locomotion, with the exception that animals with lesions as infants show some behavioral compensation in hindlimb movement by developing a normal degree of yielding at the knee. In contrast, hamsters with lesions as either adults or infants never recover normal forelimb behavior in either yielding at the elbow or accuracy of forelimb placement. These results emphasize the sensorimotor role of the pyramidal tract, even in a relatively stereotyped behavior such as locomotion. Further, they suggest that anatomical plasticity of the pyramidal tract previously reported in infant hamsters is not sufficient to preserve all normal motor behaviors, particularly accurate movements of the 0 1991 Academic Press, Inc. forelimb.
1 Present address: Department versity Medical School, 303 East
University
KALIL
Northwestern UniChicago, IL 60611.
98
EFFECTS
OF
PYRAMIDAL
TRACT
locomotion at this time (30). Of these 15 animals, 5 hamsters later received a unilateral transection of the medullary pyramid at 4 to 5 weeks of age. To eliminate possible effects of motor imbalance 4 animals received a bilateral lesion at the same age. Hamsters with pyramidal tract lesions were filmed at 5 days, 3 weeks, and 3 months following the surgery. In addition, 4 of these animals were filmed up to 1 year after the lesion. Thirteen infant golden hamsters were given a unilateral lesion of the pyramidal tract at 5 days postnatal, and their locomotor behavior was filmed 4 to 5 weeks postnatal. Surgical procedures. Adult hamsters were anesthetized with Nembutal (35 mg/kg) and infants by cooling with ice. The pyramidal tracts, which lie on the ventral surface of the medulla, were exposed by a ventral approach through the basioccipital bone and pyramidal fibers were sectioned with a scapel blade at a level just rostra1 to the inferior olivary nucleus. Since the pyramidal tract is completely crossed in hamsters (42) unilateral lesions denervate the contralateral spinal cord of all pyramidal tract inputs. Adults were returned to separate cages and infants to the mother’s nest. Filming and kinematic analysis. A 16-mm Bolex movie camera was used to film freely moving animals from a lateral view through a glass-sided, grid-backed runway approximately 4 ft long and 3 in wide. The runway was slightly curved in an arc with the camera at its focus in order to reduce parallax of the image. Adult animals and those that had reached young adulthood were filmed during locomotion on two types of terrain. Smooth terrain consisted of the flat surface of the runway, which was constructed from Styrofoam to prevent slippage of the feet. Rough terrain consisted of smooth landscaping rocks placed tightly together in the runway. The size of a 5 week old animal in relation to the terrain is shown in Fig. 1. The gait analyzed was the lateral sequence-diagonal couplet gait, or fast walk (11). Only those sequences that were smoothly performed without pauses were analyzed. Filming was carried out at 50 frames/s, and every frame or every other frame was analyzed to determine the position of the joints of the forelimb (scapula, shoulder, and elbow) and hindlimb (hip, knee, and ankle) during different phases of the step cycle. For each animal, at least 10 step cycles were analyzed on both smooth and rough terrain. Frame-by-frame movement analysis was carried out by projecting the film onto a screen from which the positions of the forelimb and hindlimb were traced. Both the hip joint and the top of the scapula were marked on the skin with black permanent ink. Analysis of the positions of the knee, shoulder, and elbow joints was aided by measurements of the length of the bones of each limb. For example, since the position of the hip and the ankle were clearly visible, the
LESIONS
ON
99
LOCOMOTION
position assigned to the knee could be checked by triangulation using the lengths of the femur and the tibia. The joint angles of the scapula, shoulder, and elbow and of the hip, knee, and ankle were measured as shown in Fig. 2. The hip angle was measured with respect to a line drawn parallel to the curve of the back just rostra1 to the hip. One hamster was X-rayed with the limbs in different positions during standing and walking to verify that this method of estimating the inclination of the pelvic girdle is accurate to within less than 5”. The scapula angle was measured with respect to a line parallel to the rostra1 part of the back. Movements of these joints could be measured in the parasagittal plane only. Joint angles were plotted against time to reveal the excursion of the limb in each step cycle. An increase in joint angle represents extension; a decrease represents flexion. Data analysis. For the purpose of analyzing the positions of the elbow, knee, and ankle in normal and lesioned animals, the step cycle was divided into four phases according to the Philippson (39) scheme: F-ElE2-E3 (see Fig. 3 and Results). The maximum degree of flexion or extension of the joints during each of these phases was recorded for step cycles within the range of walking speeds between 0.10 and 0.30 m/s. A similar analysis was made of the movements of the scapula, shoulder, and hip. Measurements of the amount of yielding of the elbow, knee, and ankle during locomotion were obtained by subtracting the peak extension in E2 from the peak extension in El (El - E2) for each step cycle. Significant differences were estimated with an analysis of variance (ANOVA), where N was the number of animals in each group. Significant difference was set at the P < 0.05 level. The numbers of animals (N) and step cycles (n) are indicated in the legend to Fig. 4. Accuracy of forelimb placement was measured on the rough terrain by observing lo-15 step cycles per animal and comparing normal limb placement (on the top surface of the rocks) with inaccurate placement (in cracks between the rocks). These observations gave the percentage accuracy of forelimb placement during locomotion. Since hindlimb placement followed the forelimb, these percentages are similar for the hindlimb as well. Histology. At the end of the behavioral analysis, the completeness of the pyramidal tract lesions in adults was assessed histologically by staining sections of the formalin-fixed brains through the lesion site with a modified Heidenhain’s stain for myelin. In the infant animals injections of horseradish peroxidase (HRP) were made in the sensorimotor cortex ipsilateral to the lesion and the brains processed by the tetramethyl benzidine method (38) to verify the completeness of the lesion. RESULTS
The normal sequence of joint coordination during locomotion of adult hamsters is stereotyped and reproduc-
100
KEIFER
AND
KALIL
FIG. 1. Photograph of a normal 5 week old hamster during locomotion on the rough terrain showing the relationship the animal to the rocky surface. The rough terrain was constructed from smooth landscaping rocks closely placed so that The background grid is 2 cm per square.
ible over both smooth and rough terrain. The pattern of joint coordination of the forelimb and hindlimb during an averaged step cycle from one typical normal animal performing locomotion over smooth terrain is illustrated in Fig. 3. As the forelimb lifts off the ground to initiate the swing phase, the scapula and elbow flex while the shoulder extends (F). This is followed by elbow extension halfway through the swing phase (El). When the forelimb is placed on the ground, the scapula extends while the shoulder flexes. There is a yield or flexion of the elbow (E2) during the stance phase when body weight is initially placed on the limb. The yield is
FIG. 2. Reconstruction of measured with respect to a line the hindlimb, the hip angle (H) were also measured as shown.
between the size of they did not wobble.
followed by a second phase of elbow extension in late stance to push off (E3). These results correspond well with previously published data of forelimb movements during locomotion of the rat (26) and the cat (15,16,24). A similar pattern of joint coordination is seen for the hindlimb (Fig. 3). As the hindlimb lifts off the ground, the hip, knee, and ankle flex together (F). Knee and ankle extension occur midway through swing (El). During stance, the hip extends while the knee and ankle yield (E2). Late stance is characterized by knee and ankle extension while the hip is extended (E3). These results from the hamster are similar to those for locomo-
a hamster skeleton showing the joint angles that were measured. For the forelimb, the scapula angle (SC) was parallel to the rostra1 portion of the back; the shoulder (Sh) and elbow (E) angles were measured as shown. For was measured with respect to a line parallel to the caudal portion of the back; the knee (K) and ankle (A) angles An increase in joint angle is extension, while a decrease is flexion.
EFFECTS
OF
PYRAMIDAL
TRACT
al v
60-
: 6040-
i,
.2’5
Time
i,-
-
.25
(set)
FIG. 3. The normal sequence of joint activity of the forelimb and hindlimb during locomotion on smooth terrain. A single step cycle is shown in each panel and is taken from the average of 10 step cycles. The movements of the scapula (SC), shoulder (Sh), and elbow (E) of the forelimb and the hip (H) knee (K) and ankle (A) of the hindlimb are plotted in joint angle in degrees. An increase in joint angle is extension, a decrease is flexion. See text for a full description of the movement. The bar represents the time the foot is on the ground during the stance phase, while the spaces represent the time the foot is off the ground and moving forward during the swing phase. The four phases of the Philippson step cycle, F-El-E2-E3, are indicated for the step cycles shown.
tion of the rat (23) and the cat (18, 22, 24) hindlimb. Coordination of the forelimb and hindlimb during locomotion over rough terrain is similar to that for locomotion over smooth terrain in normal animals despite greater postural adjustments and the need for accurate limb placement encountered on the uneven surface. Animals with either unilateral or bilateral lesions of the pyramidal tract as adults, after recovery from the surgery, readily performed locomotion through the runway. Animals with unilateral lesions showed no gross motor imbalance during locomotion. However, both bilaterally and unilaterally lesioned animals, even after several months of recovery, show differences on both smooth and rough terrain in joint movements and accuracy of limb placement during locomotion when compared to normal animals. These animals showed no behavioral recovery during the entire period of behavioral observation. As shown in Fig. 4, although the overall sequence of joint activation and the amounts of flexion and extension at the forelimb and hindlimb joints during all phases of the step cycle are essentially normal after adult pyramidal tract lesions, a marked reduction
LESIONS
ON
LOCOMOTION
101
occurs in the amount of yield at both the elbow and the knee during the step cycle. The ankle, however, remained normal. The net behavioral result is that the reduction in yielding at both the elbow and the knee joints produced an overall stiffness in both the forelimb and the hindlimb during locomotion. For the animals with adult pyramidotomy there is a statistical difference (P < 0.05) in the degree of yielding of the elbow and knee during locomotion on the smooth versus rough terrain that occurs 5 days postlesion, but this difference disappears 3 months postlesion, indicating that some compensation has occurred, especially in the hindlimb. The terrain also has striking effects on the accuracy of limb placement during locomotion. As shown in Fig. 5, normal animals show highly accurate limb placements during locomotion on rough terrain, which we define as placement of the foot on the top of the rocks during normal locomotion, as opposed to placement in the cracks between the rocks, which was inaccurate. As shown in Fig. 5, a normal animal shows close to 100% accuracy in forelimb placement, whereas the same animal after a pyramidal tract lesion declines to about 60% accuracy as measured over lo-15 step cycles. A typical step sequence by an animal with a pyramidal tract lesion began with an awkward placement of the forelimb in a crack between the rocks followed by several attempts to place the hindlimb in the same position as the inaccurately placed forelimb. Forward progression occurred at normal speed but was characterized by awkward limb placement and occasional stumbling. The deficits in locomotor behavior on both smooth and rough terrain were surprisingly similar in animals with adult or infant lesions of the pyramidal tract (Figs. 4 and 5). The most significant difference between these two groups of experimental animals occurred in the yielding phase of the hindlimb during locomotion. Whereas adult lesioned animals showed a significant reduction in the amount of yielding at the knee joint, infant lesioned animals exhibited a normal degree of yielding at this joint. Reduction of the yield phase at the elbow was similar for both groups of animals. Inaccuracies of forelimb placement during locomotion on rough terrain were also quantitatively similar for both adult and infant lesioned animals (67% versus 68% mean accuracy). The variability in motor behavior of animals that received lesions as infants or adults (e.g., Fig. 5) could not be attributed to differences in the extent of the lesion. A more likely possibility is that differing degrees of functional reorganization following the lesion occurred (see Discussion) which were not revealed with the anatomical techniques used in this study. Histology. In all of the animals with lesions as adults BO-100% of the pyramidal tract was severed (data not shown). Three animals with adult lesions also sustained
102
KEIFER
q
smooth
Ezl
rough
N
Uni
Bi
Uni
Bi
Infant
AND
N
Uni
Bi
ELBOW
FIG. 4.
KALIL
Uni
Bi
Infant
N
KNEE
Uni
Bi
Uni
Bi
Infant
ANKLE
The
amount of yield that occurs in the elbow, knee, and ankle during the step cycle is shown as the amount of extension in E2. Measurements of the degree of the yielding were obtained by subtracting maximal extension in E2 from maximal extension in El for each step cycle. Means f SEM. Step cycles of normal hamsters (N) and those that received lesions of the pyramidal tract as adults or infants (infant) performing locomotion on both smooth and rough terrain are shown. For the adult lesioned animals, the unilaterally (uni) and bilaterally (bi) lesioned groups are shown separately. Data that were obtained from animals with adult lesions 5 days (thickly outlined bars) and 3 months after the lesion are shown. Number of animals (N) and step cycles (n) represented in each bar: Normal-N = 10, n = 30,5 days postlesion, Uni-N = 5, n = 25; Bi-N = 4, n = 20; 3 months postlesion, Uni-N = 4, n = 12; Bi-N = 4, n = 12; Infant-N = 10, n = 30. *P < 0.05, ANOVA, significant differences from normal.
slight damage to the inferior olive and another three had slight damage to the medial lemniscus. However, their behavior was similar to that of animals with lesions of the pyramidal tract alone. In animals with lesions as infants, the sensorimotor cortex ipsilateral to the site of pyramidal tract lesion was injected with HRP after filming of locomotion was completed. These animals had virtually complete lesions of the pyramidal tract on one side with no detectable damage to other medullary structures. In all cases, corticofugal axons coursed through an aberrant pathway in the brain stem
and descended to the contralateral spinal cord as described by Kalil and Reh (28), innervating all cervical and lumbar segments of the cord. DISCUSSION
The major results of this study show (i) that lesions of the pyramidal tract in both infant and adult hamsters affect locomotion, particularly by a reduction in the yielding phase of the step cycle and in the accuracy of forelimb placement; (ii) that locomotion on rough
r aQ
ABCDEGHI Normal
JKLMNOPQRBTUV
ABDEFGHI 3 months
post-lesion
Infant
lesion
FIG. 5. Histogram showing the percentage accuracy of forelimb placement of normal animals and those with pyramidal tract lesions during locomotion on rough terrain. Each animal is represented by a letter. Adult lesioned animals with unilateral lesions are illustrated with white bars; those with bilateral lesions are illustrated with hatched bars. For the data from normal and infant lesioned animals, 15 step cycles of each animal were examined for accuracy of forelimb placement. Ten steps per animal were examined 3 months following the lesion in adult animals. Accurate placement occured when the foot was placed on top of a rock, while inaccurate placement occured when the foot was placed in a crack between the rocks or awkwardly along the steep side of a rock. Note that animal C could not be tested 3 months postlesion and that normal data from animal F was not available. The mean percentage accuracy for all the subjects within the three groups is indicated to the left: open triangle, normal adults (91%); closed diamond, adults 3 months postlesion (67%); open diamond, infant lesion (68%).
EFFECTS
OF
PYRAMIDAL
TRACT
terrain accentuates placing deficits; and (iii) that, while infant-lesioned animals show some behavioral compensation, i.e., by developing a normal degree of yielding in the hindlimb, they are unable to develop normal yielding and accuracy of forelimb movements during locomotion. Previous studies on the functional role of the pyramidal tract in various species including rodents (13,25,27, 43), cats (al), and primates (20,34,35) have emphasized its participation in fine motor control of the distal forelimb. Anatomically, the dual origins of the primate (10, 12, 41) pyramidal tract from somatosensory and motor cortex and its innervation of the dorsal and ventral horns, respectively, suggest a dual sensory and motor function of its corticospinal components. Recently, anatomical studies of the hamster pyramidal tract revealed similar differential sensory and motor projections and also demonstrated direct connections to the ventral motorneuron pools (33) within the cervical and lumbar enlargements. These anatomical findings suggest that lesions of the pyramidal tract in hamsters, as in primates, may therefore interfere with two aspects of movements of both the forelimb and the hindlimb, i.e., the modulation of sensory input to the dorsal horn and the direct activation of ventral horn neurons involved in organizing movement. Physiological studies using an in vitro hamster spinal cord preparation (29) support the notion that descending corticospinal fibers modulate sensory input by exerting an inhibitory effect on dorsal root afferents (17). Therefore, given the evidence that the pyramidal tract in rodents plays a role in both the sensory and the motor components of movement it is perhaps not surprising that interruption of this pathway should, in addition to previously observed disruption of fine manipulatory movements of the forepaw, also disrupt certain components of locomotion, even though this is a relatively stereotyped motor behavior. Previous studies have reported that animals with cortical or pyramidal tract lesions appeared to move stiffly and the limbs showed an increased resistance to passive flexion (2, 8, 14,36). Further, previous reports on the effects of pyramidal tract lesions on placing responses in rodents (13, 25) are consistent with the deficits in accuracy of limb placement during locomotion observed in the present study. It is not entirely clear why the yield phase of the step cycle is particularly dependent on intact pyramidal tract innervation. One possibility is that pyramidotomy disrupts normal influences on stretch reflexes (9, 19, 29, 40). In light of the important sensorimotor function of the pyramidal tract, it is also not surprising that lesions of this pathway have more noticeable effects on locomotion performed on rough terrain as opposed to a smooth surface. Joint movements and foot placements on the rocky terrain require continual sensory feedback from
LESIONS
ON
LOCOMOTION
103
the periphery in order for the animal to make limb and postural adjustments. Moreover, the rocky terrain requires accurate limb placing movements, which are dependent on an intact descending corticospinal influence on interneurons and motorneurons activating functionally related muscles. In contrast, locomotion on smooth terrain would be more automatic and dependent on local spinal circuitry (1). An unexpected result of the present study was the similarity between adult and infant lesioned animals in their locomotor behavior. Given the greater anatomical plasticity of the pyramidal tract in young rodents (3,4, 28) than in adults (5, 6, 7), one would have predicted a large difference between infant and adult operates in recovery of motor function. Previous studies of the effects of pyamidotomy on motor behavior showed that hamsters with lesions as adults never recovered fine manipulatory movements of the forelimb and digits (27, 43), whereas animals with lesions as infants were able to develop normal fine motor control of the forelimb (43). It should be noted that these studies were largely descriptive and measured the time required for the animals to shell sunflower seeds, a skilled forelimb behavior. Moreover, careful observations of filmed motor behavior revealed some loss of manual dexterity in the infant lesioned adult hamsters. Thus, the present results, showing that animals with pyramidal tract lesions at 5 days postnatal do not fully recover normal yielding or placing of the forelimb during locomotion, are consistent with previous results in demonstrating that forelimb movement is especially dependent on intact pyramidal tract connections and is less able to compensate for the effects of early pyramidotomy than hindlimb movement. The possibility remains, however, that deficits in forelimb movements of neonatal operates might improve with age, after the postoperative period in which motor behavior was examined in this study. The difference between the ability of the forelimb and hindlimb to develop a normal degree of yielding after early pyramidotomy is somewhat surprising from an anatomical point of view. Cortical injections of HRP ipsilateral to the site of a neonatal lesion of the pyramidal tract reveals aberrant corticospinal projections to both the cervical and the lumbar spinal cord. Furthermore, recent studies have shown that, although axotomized corticospinal neurons die after pyramidotomy at 5 days postnatal (37), regions of the contralateral cord are subsequently invaded by numerous corticospinal axons sprouting from the normal side of the cord (32). This sprouting declines sharply with age. Other types of reorganization from local or descending pathways may also occur in animals with infant pyramidal tract lesions. Since the sprouting axons form corticospinal arbors that are topographically specific and that show normal morphology in both the cervical and the lumbar enlarge-
104
KEIFER
AND
ments, it is puzzling that this precise anatomical reorganization cannot preserve normal motor control of the forelimb. This finding leads to the conclusion that restoration of appropriate connectivity by redirected growth of axons ipsilateral to the lesion site, or by compensatory sprouting from the opposite side of the cord, does not necessarily preserve normal motor function under all conditions. Further, it stresses the particular importance of the pyramidal tract in motor behavior of the forelimb which may require precise patterns of corticospinal innervation for completely normal movement. ACKNOWLEDGMENTS We thank Laurel Carney, James C. Houk, Carolyn Norris, Dean 0. Smith, and Paul S. G. Stein for helpful comments on earlier versions of the manuscript. We also thank Paul Stein for valuable discussions during the course of this work and Cheryl Adams for reconstruction of the hamster skeleton and preparation of the figures. This research was supported by NSF Grant BNS-8311517 and NIH Grant NS 14428 to K.K.
AMSTRONG, D. M. 1986. Supraspinal contributions tion and control of locomotion in the cat. Prog.
14.
to the initiaNeurobiol. 26:
15’ re.
3.
D. in the rat. BERNSTEIN, corticospinal postnatal
H. 1934. The results of unilateral pyramidal J. Comp. Neural. 60: 45-56. D. R., AND D. J. STELZNER. 1983. Plasticity tract following midthoracic spinal injury rat. J. Comp. Neural. 221: 382-400.
section of the in the
4. BREGMAN,
B. S., E. KUNKEL-BAGDEN, M. MCATEE, AND A. O’NEILL. 1989. Extension of the critical period for developmental plasticity of the corticospinal pathway. J. Comp. Neural.
282:355-370.
76:105-122.
20.
CHEEMA, S. S., A. RUSTIONI, AND B. L. WHITSEL. 1984. Light and electron microscopic evidence for a direct corticospinal projection to superficial laminae of the dorsal horn in cats and monkeys. J. Comp. Neurol. 225: 276-290.
11.
COHEN, A. H., AND C. GANS. 1975. Muscle activity in rat locomotion: Movement analysis and electromyography of the flexors and extensors of the elbow. J. Morphol. 146: 177-196.
12.
COULTER, J. D., AND E. G. JONES. 1977. Differential distribution of corticospinal projections from individual cytoarchitectonic fields in the monkey. Brain Res. 129: 335-340.
13.
DONATELLE,
J. M.
197’7. Growth
of the corticospinal
tract
The dur-
23.
of CinRes.
G. E., R. M. REINKING, AND D. G. STUART. 1973. cat step cycle: Hind limb joint angles and muscle lengths ing unrestrained locomotion. J. Morphol. 141: l-42. GRUNER, J. A., J. ALTMAN, AND N. SPIVACK. 1980. Effects arrested cerebellar development on locomotion in the rat: ematographic and electromyographic analysis. Ezp. Bruin
40:361-373. 24. HALBERTSMA,
J. M. 1983. The stride cycle of the cat: The modelling of locomotion by computerized analysis of automatic recordings. Acta Physiol. Stand. (Suppl.) 521: l-75.
25. HICKS,
S. P., AND C. J. D’AMATO. 1975. Motor-sensory cortexcorticospinal system and developing locomotion and placing in rats. Am. J. Anat. 143: l-42.
26. JENKINS,
F. A. 1974. The movement of the shoulder late and aclaviculate mammals. J. Morphol. 144:
in clavicu-
71-84.
K., AND G. E. SCHNEIDER. 1975. Motor performance lowing unilateral pyramidal tract lesions in the hamster. Res. 100: 170-174.
study
K., AND T. REH. 1982. A light of regrowing pyramidal tract
folBruin
and electron microscopic fibers. J. Comp. Neurol.
211:265-275. 29. KEIFER, J., AND K. KALIL. pyramidal tract cord preparation
640. 30. KEIFER, 31.
J., hamsters Neurosci. KEIFER, J., pyramidal 1030.
1989. Modulation of spinal reflexes by stimulation in an in uitro brainstem-spinal from the hamster. Exp. Bruin Res. 78: 633-
AND K. KALIL. 1987. Development of locomotion in with neonatal lesions of the pyramidal tract. Sot. Abstr. 13: 1503. AND K. KALIL. 1985. Functional role of the hamster tract during locomotion. Sot. Neurosci. Abstr. 11:
32. KUANG, 33.
R. Z., AND K. KALIL. 1989. Sprouting of corticospinal fibers into denervated contralateral spinal cord reveals specificity of new axon arbors. Sot. Neurosci. Abstr. 15: 92. KUANG, R. Z., AND K. KALIL. 1990. Branching patterns of corticospinal axon arbors in the rodent. J. Comp. Neurol., in press.
34. LAWRENCE, and
of pyramidal lesions Neurobiol. Exp. 40:
22. GOSLOW,
387. 10.
in the
31: 69-80.
AND J. HALBERTSMA. 1980. The cat. I. Coordination within a hind108: 269-281.
GORSKA, T., AND E. SYEURSKA. 1980. Effects on forelimb movements in the cat. Acta 843-859.
28. KALIL,
CHAPMAN, C. E., AND M. WIESENDANGER. 1982. Recovery of function following unilateral lesions of the bulbar pyramid in the monkey. Electroencephnlogr. Clin. Neurophysiol. 53: 374-
on interneurons
GILMAN, S., L. A. MARCO, AND H. C. EBEL. 1971. Effects of medullary pyramidotomy in the monkey. II. Abnormalities of spindle responses. Brain 94: 515-530. GOLDBERGER, M. E. 1969. The extrapyramidal systems of the spinal cord. II. Results of combined pyramidal and extrapyramidal lesions in the macaque. J. Comp. Neurol. 135: l-26.
BREGMAN, B. S., AND M. E. GOLDBERGER. 1983. Infant lesion effect. III. Anatomical correlates of sparing and recovery of function after spinal cord damage in newborn and adult cats. Dev. Brain Res. 9: 137-154.
9.
tract effects J. Neurophysiol.
19.
7.
of
horn.
girdle
FORSSBERG, H., S. GRILLNER, locomotion of the low spinal limb. Acta Physiol. Scund.
27. KALIL,
tract
dorsal
of the shoulder
166: 279-292.
18.
B. S., AND M. E. GOLDBERGER. 1983. Infant lesion effect. II. Sparing and recovery of function after spinal cord damage in newborn and adult cats. Deu. Bruin Res. 9: 119-135.
CHAMBERS, W. W., AND C.-N. LIU. 1957. Cortico-spinal the cat. J. Comp. Neurol. 108: 23-56.
analysis of forelimb in the cat. J. Exp. Biol.
ENGLISH, A. W. 1978. Functional analysis of cats during locomotion. J. Morphol.
6. BREGMAN,
8.
A. W. 1978. An electromyographic during overground stepping
cat lumbar
5. BREGMAN,
B. S., AND M. E. GOLDBERGER. 1983. Infant lesion effect. I. Effect of neonatal spinal cord lesion on motor development in the kitten. Deu. Brain Res. 9: 103-117.
ENGLISH, muscles
17. FETZ, E. E. 1968. Pyramidal
273-361. 2. BARRON,
the development of placing reactions in the postnatal rat. J. Camp. Neural. 176:207-232. EIDELBERG, E., AND J. Yu. 1981. Effects of corticospinal lesions upon treadmill locomotion by cats. Exp. Bruin Res. 43: lOl-
103.
21.
REFERENCES 1.
KALIL
D. G., AND H. G. J. M. KUYPERS. 1968. The functional organization of the motor system in the monkey. I. The effects of bilateral pyramidal lesions. Bruin 91: 1-14.
EFFECTS
OF
PYRAMIDAL
35.
LEMON, R. 1988. The output map of the primate Trends Neurosci. 11: 501-506.
36.
LIDDELL, E. G. T., AND C. G. PHILLIPS. in the cat. Brain 67: 1-9.
37.
MERLINE, M., AND K. KALIL. 1990. Cell death of corticospinal neurons is induced by axotomy before but not after innervation of spinal targets. J. Comp. Neural. 296: 506-516. MESULAM, M. 1978. Tetramethyl benzidine for horseradish peroxidase neurochemistry: A non-carcinogenic blue reactionproduct with superior sensitivity for visualizing neural afferents and efferents. J. Histochem. Cytochem. 26: 106-117. PHILIPPSON, M. 1905. L’autonomie et la centralization dans le systeme nerveux des animaux. Trav. Lab. Physiol. Inst. Soluay. 7: l-208.
38.
39.
motor
TRACT
1944. Pyramidal
cortex.
LESIONS
ON
LOCOMOTION
105
40.
PRESTON, J. B., M. C. SHENDE, AND K. UEMURA. 1967. The motor cortex-pyramidal system: Pattern of facilitation andinhibition on motoneurons innervating limb musculature of the cat and baboon and their possible adaptive significance. In Neurophysiological Basis of Normal and Abnormal Activities (M. D. Yahr and D. P. Purpura, Eds.), pp. 61-72. Raven, New York.
41.
RALSTON, D. D., AND H. J. RALSTON. 1985. The terminations of corticospinal tract axons in the macaque monkey. J. Comp. Neurol. 242: 325-337.
42.
REH, T., AND K. KALIL. 1981. Development tract in the hamster. I. A light microscopic Neural. 200: 55-67.
43.
REH, T., AND K. KALIL. 1982. Functional midal tract fibers. J. Comp. Neural.
section
of the pyramidal study. J. Camp.
role of regrowing
211: 176-283.
pyra-
NEUROLOGY
111,98-105
(19%)
Effects of Infant versus Adult Pyramidal Tract Lesions on Locomotor Behavior in Hamsters JOYCE KEIFER’ Neuroscience
Training
Program
and Department
AND KATHERINE of Anatomy,
MATERIALS
INTRODUCTION
0014.4886/91
$3.00
Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
Madison,
Wzkconsin
53706
AND
METHODS
Fifteen normal adult golden hamsters (Mesocricetus aurutus) ranging in age from 3 weeks to 6 months were filmed as they performed locomotion. Anatomically, the pyramidal tract of hamsters is well developed by 3 weeks of age (42) and hamsters show a mature form of
A major descending pathway, the pyramidal tract, arises from the sensorimotor cortex and in most mamof Physiology, Chicago Avenue,
of Wisconsin,
mals projects to all segments of the spinal cord. Previous studies of the pyramidal tract have demonstrated its importance in behaviors that require a high degree of motor skill, particularly during fine manipulatory movements of the forepaw and digits. For example, after a lesion of the medullary pyramids, monkeys are unable to use the digits independently (34). Rodents with a pyramidal tract lesion exhibit deficits in placing (13,25) and fine control of the forepaw (25,27,43). Cats with pyramidal tract lesions are also impaired in forelimb reaching movements (21). However, because of its extensive projections to the dorsal horn of the spinal cord, the pyramidal tract may also play a more general role in modulating behaviors requiring sensory feedback. Thus, even those behaviors that are relatively stereotyped, such as locomotion, may be influenced by descending corticospinal input. Several studies suggest that the pyramidal tract does play a role in locomotion. Observations of animals with pyramidal tract lesions reveal disorders in posture and gait, especially when skill beyond the mere rhythm of locomotion is required (2, 8, 14, 25, 36). The aims of the present study were twofold. First, given the evidence that the pyramidal tract is required for skilled movements involving sensory feedback and that it may also play a role in locomotion, we wished to determine how lesions of the pyramidal tract might disrupt joint movements and foot placements during locomotion in a situation requiring continual adjustment to a changing terrain. The second goal was to determine whether animals with lesions as infants, in which anatomical rearrangements are known to occur after pyramidal tract lesions (3, 4, 28, 32), would show greater functional recovery in locomotor behavior than animals with adult pyramidotomy. Some of the results have been previously reported (30, 31).
The role of the pyramidal tract in locomotion was studied in hamsters by analyzing their locomotor behavior after lesions of the medullary pyramidal tract. Animals with lesions either as adults or as infants were compared to determine whether early pyramidotomy results in greater functional recovery. Normal and pyramidotomized animals were filmed during locomotion on a runway consisting of either smooth or rough terrain to assess whether the uneven surface would accentuate locomotor deficits. Frame-by-frame analysis of the filmed behavior during all phases of the step cycle was carried out to determine positions of the joints of the forelimb and hindlimb during locomotion. Accuracy of limb placement on the rough terrain was determined by observations of consecutive step cycles. The results show that lesions of the pyramidal tract in both infant and adult hamsters affect locomotion first by causing a reduction in the yielding phase of the step cycle and second by producing inaccuracies of forelimb placement. Rough terrain accentuates deficits in forelimb placement during locomotion. Animals with lesions as infants and those with lesions as adults show surprisingly similar deficits in locomotion, with the exception that animals with lesions as infants show some behavioral compensation in hindlimb movement by developing a normal degree of yielding at the knee. In contrast, hamsters with lesions as either adults or infants never recover normal forelimb behavior in either yielding at the elbow or accuracy of forelimb placement. These results emphasize the sensorimotor role of the pyramidal tract, even in a relatively stereotyped behavior such as locomotion. Further, they suggest that anatomical plasticity of the pyramidal tract previously reported in infant hamsters is not sufficient to preserve all normal motor behaviors, particularly accurate movements of the 0 1991 Academic Press, Inc. forelimb.
1 Present address: Department versity Medical School, 303 East
University
KALIL
Northwestern UniChicago, IL 60611.
98
EFFECTS
OF
PYRAMIDAL
TRACT
locomotion at this time (30). Of these 15 animals, 5 hamsters later received a unilateral transection of the medullary pyramid at 4 to 5 weeks of age. To eliminate possible effects of motor imbalance 4 animals received a bilateral lesion at the same age. Hamsters with pyramidal tract lesions were filmed at 5 days, 3 weeks, and 3 months following the surgery. In addition, 4 of these animals were filmed up to 1 year after the lesion. Thirteen infant golden hamsters were given a unilateral lesion of the pyramidal tract at 5 days postnatal, and their locomotor behavior was filmed 4 to 5 weeks postnatal. Surgical procedures. Adult hamsters were anesthetized with Nembutal (35 mg/kg) and infants by cooling with ice. The pyramidal tracts, which lie on the ventral surface of the medulla, were exposed by a ventral approach through the basioccipital bone and pyramidal fibers were sectioned with a scapel blade at a level just rostra1 to the inferior olivary nucleus. Since the pyramidal tract is completely crossed in hamsters (42) unilateral lesions denervate the contralateral spinal cord of all pyramidal tract inputs. Adults were returned to separate cages and infants to the mother’s nest. Filming and kinematic analysis. A 16-mm Bolex movie camera was used to film freely moving animals from a lateral view through a glass-sided, grid-backed runway approximately 4 ft long and 3 in wide. The runway was slightly curved in an arc with the camera at its focus in order to reduce parallax of the image. Adult animals and those that had reached young adulthood were filmed during locomotion on two types of terrain. Smooth terrain consisted of the flat surface of the runway, which was constructed from Styrofoam to prevent slippage of the feet. Rough terrain consisted of smooth landscaping rocks placed tightly together in the runway. The size of a 5 week old animal in relation to the terrain is shown in Fig. 1. The gait analyzed was the lateral sequence-diagonal couplet gait, or fast walk (11). Only those sequences that were smoothly performed without pauses were analyzed. Filming was carried out at 50 frames/s, and every frame or every other frame was analyzed to determine the position of the joints of the forelimb (scapula, shoulder, and elbow) and hindlimb (hip, knee, and ankle) during different phases of the step cycle. For each animal, at least 10 step cycles were analyzed on both smooth and rough terrain. Frame-by-frame movement analysis was carried out by projecting the film onto a screen from which the positions of the forelimb and hindlimb were traced. Both the hip joint and the top of the scapula were marked on the skin with black permanent ink. Analysis of the positions of the knee, shoulder, and elbow joints was aided by measurements of the length of the bones of each limb. For example, since the position of the hip and the ankle were clearly visible, the
LESIONS
ON
99
LOCOMOTION
position assigned to the knee could be checked by triangulation using the lengths of the femur and the tibia. The joint angles of the scapula, shoulder, and elbow and of the hip, knee, and ankle were measured as shown in Fig. 2. The hip angle was measured with respect to a line drawn parallel to the curve of the back just rostra1 to the hip. One hamster was X-rayed with the limbs in different positions during standing and walking to verify that this method of estimating the inclination of the pelvic girdle is accurate to within less than 5”. The scapula angle was measured with respect to a line parallel to the rostra1 part of the back. Movements of these joints could be measured in the parasagittal plane only. Joint angles were plotted against time to reveal the excursion of the limb in each step cycle. An increase in joint angle represents extension; a decrease represents flexion. Data analysis. For the purpose of analyzing the positions of the elbow, knee, and ankle in normal and lesioned animals, the step cycle was divided into four phases according to the Philippson (39) scheme: F-ElE2-E3 (see Fig. 3 and Results). The maximum degree of flexion or extension of the joints during each of these phases was recorded for step cycles within the range of walking speeds between 0.10 and 0.30 m/s. A similar analysis was made of the movements of the scapula, shoulder, and hip. Measurements of the amount of yielding of the elbow, knee, and ankle during locomotion were obtained by subtracting the peak extension in E2 from the peak extension in El (El - E2) for each step cycle. Significant differences were estimated with an analysis of variance (ANOVA), where N was the number of animals in each group. Significant difference was set at the P < 0.05 level. The numbers of animals (N) and step cycles (n) are indicated in the legend to Fig. 4. Accuracy of forelimb placement was measured on the rough terrain by observing lo-15 step cycles per animal and comparing normal limb placement (on the top surface of the rocks) with inaccurate placement (in cracks between the rocks). These observations gave the percentage accuracy of forelimb placement during locomotion. Since hindlimb placement followed the forelimb, these percentages are similar for the hindlimb as well. Histology. At the end of the behavioral analysis, the completeness of the pyramidal tract lesions in adults was assessed histologically by staining sections of the formalin-fixed brains through the lesion site with a modified Heidenhain’s stain for myelin. In the infant animals injections of horseradish peroxidase (HRP) were made in the sensorimotor cortex ipsilateral to the lesion and the brains processed by the tetramethyl benzidine method (38) to verify the completeness of the lesion. RESULTS
The normal sequence of joint coordination during locomotion of adult hamsters is stereotyped and reproduc-
100
KEIFER
AND
KALIL
FIG. 1. Photograph of a normal 5 week old hamster during locomotion on the rough terrain showing the relationship the animal to the rocky surface. The rough terrain was constructed from smooth landscaping rocks closely placed so that The background grid is 2 cm per square.
ible over both smooth and rough terrain. The pattern of joint coordination of the forelimb and hindlimb during an averaged step cycle from one typical normal animal performing locomotion over smooth terrain is illustrated in Fig. 3. As the forelimb lifts off the ground to initiate the swing phase, the scapula and elbow flex while the shoulder extends (F). This is followed by elbow extension halfway through the swing phase (El). When the forelimb is placed on the ground, the scapula extends while the shoulder flexes. There is a yield or flexion of the elbow (E2) during the stance phase when body weight is initially placed on the limb. The yield is
FIG. 2. Reconstruction of measured with respect to a line the hindlimb, the hip angle (H) were also measured as shown.
between the size of they did not wobble.
followed by a second phase of elbow extension in late stance to push off (E3). These results correspond well with previously published data of forelimb movements during locomotion of the rat (26) and the cat (15,16,24). A similar pattern of joint coordination is seen for the hindlimb (Fig. 3). As the hindlimb lifts off the ground, the hip, knee, and ankle flex together (F). Knee and ankle extension occur midway through swing (El). During stance, the hip extends while the knee and ankle yield (E2). Late stance is characterized by knee and ankle extension while the hip is extended (E3). These results from the hamster are similar to those for locomo-
a hamster skeleton showing the joint angles that were measured. For the forelimb, the scapula angle (SC) was parallel to the rostra1 portion of the back; the shoulder (Sh) and elbow (E) angles were measured as shown. For was measured with respect to a line parallel to the caudal portion of the back; the knee (K) and ankle (A) angles An increase in joint angle is extension, while a decrease is flexion.
EFFECTS
OF
PYRAMIDAL
TRACT
al v
60-
: 6040-
i,
.2’5
Time
i,-
-
.25
(set)
FIG. 3. The normal sequence of joint activity of the forelimb and hindlimb during locomotion on smooth terrain. A single step cycle is shown in each panel and is taken from the average of 10 step cycles. The movements of the scapula (SC), shoulder (Sh), and elbow (E) of the forelimb and the hip (H) knee (K) and ankle (A) of the hindlimb are plotted in joint angle in degrees. An increase in joint angle is extension, a decrease is flexion. See text for a full description of the movement. The bar represents the time the foot is on the ground during the stance phase, while the spaces represent the time the foot is off the ground and moving forward during the swing phase. The four phases of the Philippson step cycle, F-El-E2-E3, are indicated for the step cycles shown.
tion of the rat (23) and the cat (18, 22, 24) hindlimb. Coordination of the forelimb and hindlimb during locomotion over rough terrain is similar to that for locomotion over smooth terrain in normal animals despite greater postural adjustments and the need for accurate limb placement encountered on the uneven surface. Animals with either unilateral or bilateral lesions of the pyramidal tract as adults, after recovery from the surgery, readily performed locomotion through the runway. Animals with unilateral lesions showed no gross motor imbalance during locomotion. However, both bilaterally and unilaterally lesioned animals, even after several months of recovery, show differences on both smooth and rough terrain in joint movements and accuracy of limb placement during locomotion when compared to normal animals. These animals showed no behavioral recovery during the entire period of behavioral observation. As shown in Fig. 4, although the overall sequence of joint activation and the amounts of flexion and extension at the forelimb and hindlimb joints during all phases of the step cycle are essentially normal after adult pyramidal tract lesions, a marked reduction
LESIONS
ON
LOCOMOTION
101
occurs in the amount of yield at both the elbow and the knee during the step cycle. The ankle, however, remained normal. The net behavioral result is that the reduction in yielding at both the elbow and the knee joints produced an overall stiffness in both the forelimb and the hindlimb during locomotion. For the animals with adult pyramidotomy there is a statistical difference (P < 0.05) in the degree of yielding of the elbow and knee during locomotion on the smooth versus rough terrain that occurs 5 days postlesion, but this difference disappears 3 months postlesion, indicating that some compensation has occurred, especially in the hindlimb. The terrain also has striking effects on the accuracy of limb placement during locomotion. As shown in Fig. 5, normal animals show highly accurate limb placements during locomotion on rough terrain, which we define as placement of the foot on the top of the rocks during normal locomotion, as opposed to placement in the cracks between the rocks, which was inaccurate. As shown in Fig. 5, a normal animal shows close to 100% accuracy in forelimb placement, whereas the same animal after a pyramidal tract lesion declines to about 60% accuracy as measured over lo-15 step cycles. A typical step sequence by an animal with a pyramidal tract lesion began with an awkward placement of the forelimb in a crack between the rocks followed by several attempts to place the hindlimb in the same position as the inaccurately placed forelimb. Forward progression occurred at normal speed but was characterized by awkward limb placement and occasional stumbling. The deficits in locomotor behavior on both smooth and rough terrain were surprisingly similar in animals with adult or infant lesions of the pyramidal tract (Figs. 4 and 5). The most significant difference between these two groups of experimental animals occurred in the yielding phase of the hindlimb during locomotion. Whereas adult lesioned animals showed a significant reduction in the amount of yielding at the knee joint, infant lesioned animals exhibited a normal degree of yielding at this joint. Reduction of the yield phase at the elbow was similar for both groups of animals. Inaccuracies of forelimb placement during locomotion on rough terrain were also quantitatively similar for both adult and infant lesioned animals (67% versus 68% mean accuracy). The variability in motor behavior of animals that received lesions as infants or adults (e.g., Fig. 5) could not be attributed to differences in the extent of the lesion. A more likely possibility is that differing degrees of functional reorganization following the lesion occurred (see Discussion) which were not revealed with the anatomical techniques used in this study. Histology. In all of the animals with lesions as adults BO-100% of the pyramidal tract was severed (data not shown). Three animals with adult lesions also sustained
102
KEIFER
q
smooth
Ezl
rough
N
Uni
Bi
Uni
Bi
Infant
AND
N
Uni
Bi
ELBOW
FIG. 4.
KALIL
Uni
Bi
Infant
N
KNEE
Uni
Bi
Uni
Bi
Infant
ANKLE
The
amount of yield that occurs in the elbow, knee, and ankle during the step cycle is shown as the amount of extension in E2. Measurements of the degree of the yielding were obtained by subtracting maximal extension in E2 from maximal extension in El for each step cycle. Means f SEM. Step cycles of normal hamsters (N) and those that received lesions of the pyramidal tract as adults or infants (infant) performing locomotion on both smooth and rough terrain are shown. For the adult lesioned animals, the unilaterally (uni) and bilaterally (bi) lesioned groups are shown separately. Data that were obtained from animals with adult lesions 5 days (thickly outlined bars) and 3 months after the lesion are shown. Number of animals (N) and step cycles (n) represented in each bar: Normal-N = 10, n = 30,5 days postlesion, Uni-N = 5, n = 25; Bi-N = 4, n = 20; 3 months postlesion, Uni-N = 4, n = 12; Bi-N = 4, n = 12; Infant-N = 10, n = 30. *P < 0.05, ANOVA, significant differences from normal.
slight damage to the inferior olive and another three had slight damage to the medial lemniscus. However, their behavior was similar to that of animals with lesions of the pyramidal tract alone. In animals with lesions as infants, the sensorimotor cortex ipsilateral to the site of pyramidal tract lesion was injected with HRP after filming of locomotion was completed. These animals had virtually complete lesions of the pyramidal tract on one side with no detectable damage to other medullary structures. In all cases, corticofugal axons coursed through an aberrant pathway in the brain stem
and descended to the contralateral spinal cord as described by Kalil and Reh (28), innervating all cervical and lumbar segments of the cord. DISCUSSION
The major results of this study show (i) that lesions of the pyramidal tract in both infant and adult hamsters affect locomotion, particularly by a reduction in the yielding phase of the step cycle and in the accuracy of forelimb placement; (ii) that locomotion on rough
r aQ
ABCDEGHI Normal
JKLMNOPQRBTUV
ABDEFGHI 3 months
post-lesion
Infant
lesion
FIG. 5. Histogram showing the percentage accuracy of forelimb placement of normal animals and those with pyramidal tract lesions during locomotion on rough terrain. Each animal is represented by a letter. Adult lesioned animals with unilateral lesions are illustrated with white bars; those with bilateral lesions are illustrated with hatched bars. For the data from normal and infant lesioned animals, 15 step cycles of each animal were examined for accuracy of forelimb placement. Ten steps per animal were examined 3 months following the lesion in adult animals. Accurate placement occured when the foot was placed on top of a rock, while inaccurate placement occured when the foot was placed in a crack between the rocks or awkwardly along the steep side of a rock. Note that animal C could not be tested 3 months postlesion and that normal data from animal F was not available. The mean percentage accuracy for all the subjects within the three groups is indicated to the left: open triangle, normal adults (91%); closed diamond, adults 3 months postlesion (67%); open diamond, infant lesion (68%).
EFFECTS
OF
PYRAMIDAL
TRACT
terrain accentuates placing deficits; and (iii) that, while infant-lesioned animals show some behavioral compensation, i.e., by developing a normal degree of yielding in the hindlimb, they are unable to develop normal yielding and accuracy of forelimb movements during locomotion. Previous studies on the functional role of the pyramidal tract in various species including rodents (13,25,27, 43), cats (al), and primates (20,34,35) have emphasized its participation in fine motor control of the distal forelimb. Anatomically, the dual origins of the primate (10, 12, 41) pyramidal tract from somatosensory and motor cortex and its innervation of the dorsal and ventral horns, respectively, suggest a dual sensory and motor function of its corticospinal components. Recently, anatomical studies of the hamster pyramidal tract revealed similar differential sensory and motor projections and also demonstrated direct connections to the ventral motorneuron pools (33) within the cervical and lumbar enlargements. These anatomical findings suggest that lesions of the pyramidal tract in hamsters, as in primates, may therefore interfere with two aspects of movements of both the forelimb and the hindlimb, i.e., the modulation of sensory input to the dorsal horn and the direct activation of ventral horn neurons involved in organizing movement. Physiological studies using an in vitro hamster spinal cord preparation (29) support the notion that descending corticospinal fibers modulate sensory input by exerting an inhibitory effect on dorsal root afferents (17). Therefore, given the evidence that the pyramidal tract in rodents plays a role in both the sensory and the motor components of movement it is perhaps not surprising that interruption of this pathway should, in addition to previously observed disruption of fine manipulatory movements of the forepaw, also disrupt certain components of locomotion, even though this is a relatively stereotyped motor behavior. Previous studies have reported that animals with cortical or pyramidal tract lesions appeared to move stiffly and the limbs showed an increased resistance to passive flexion (2, 8, 14,36). Further, previous reports on the effects of pyramidal tract lesions on placing responses in rodents (13, 25) are consistent with the deficits in accuracy of limb placement during locomotion observed in the present study. It is not entirely clear why the yield phase of the step cycle is particularly dependent on intact pyramidal tract innervation. One possibility is that pyramidotomy disrupts normal influences on stretch reflexes (9, 19, 29, 40). In light of the important sensorimotor function of the pyramidal tract, it is also not surprising that lesions of this pathway have more noticeable effects on locomotion performed on rough terrain as opposed to a smooth surface. Joint movements and foot placements on the rocky terrain require continual sensory feedback from
LESIONS
ON
LOCOMOTION
103
the periphery in order for the animal to make limb and postural adjustments. Moreover, the rocky terrain requires accurate limb placing movements, which are dependent on an intact descending corticospinal influence on interneurons and motorneurons activating functionally related muscles. In contrast, locomotion on smooth terrain would be more automatic and dependent on local spinal circuitry (1). An unexpected result of the present study was the similarity between adult and infant lesioned animals in their locomotor behavior. Given the greater anatomical plasticity of the pyramidal tract in young rodents (3,4, 28) than in adults (5, 6, 7), one would have predicted a large difference between infant and adult operates in recovery of motor function. Previous studies of the effects of pyamidotomy on motor behavior showed that hamsters with lesions as adults never recovered fine manipulatory movements of the forelimb and digits (27, 43), whereas animals with lesions as infants were able to develop normal fine motor control of the forelimb (43). It should be noted that these studies were largely descriptive and measured the time required for the animals to shell sunflower seeds, a skilled forelimb behavior. Moreover, careful observations of filmed motor behavior revealed some loss of manual dexterity in the infant lesioned adult hamsters. Thus, the present results, showing that animals with pyramidal tract lesions at 5 days postnatal do not fully recover normal yielding or placing of the forelimb during locomotion, are consistent with previous results in demonstrating that forelimb movement is especially dependent on intact pyramidal tract connections and is less able to compensate for the effects of early pyramidotomy than hindlimb movement. The possibility remains, however, that deficits in forelimb movements of neonatal operates might improve with age, after the postoperative period in which motor behavior was examined in this study. The difference between the ability of the forelimb and hindlimb to develop a normal degree of yielding after early pyramidotomy is somewhat surprising from an anatomical point of view. Cortical injections of HRP ipsilateral to the site of a neonatal lesion of the pyramidal tract reveals aberrant corticospinal projections to both the cervical and the lumbar spinal cord. Furthermore, recent studies have shown that, although axotomized corticospinal neurons die after pyramidotomy at 5 days postnatal (37), regions of the contralateral cord are subsequently invaded by numerous corticospinal axons sprouting from the normal side of the cord (32). This sprouting declines sharply with age. Other types of reorganization from local or descending pathways may also occur in animals with infant pyramidal tract lesions. Since the sprouting axons form corticospinal arbors that are topographically specific and that show normal morphology in both the cervical and the lumbar enlarge-
104
KEIFER
AND
ments, it is puzzling that this precise anatomical reorganization cannot preserve normal motor control of the forelimb. This finding leads to the conclusion that restoration of appropriate connectivity by redirected growth of axons ipsilateral to the lesion site, or by compensatory sprouting from the opposite side of the cord, does not necessarily preserve normal motor function under all conditions. Further, it stresses the particular importance of the pyramidal tract in motor behavior of the forelimb which may require precise patterns of corticospinal innervation for completely normal movement. ACKNOWLEDGMENTS We thank Laurel Carney, James C. Houk, Carolyn Norris, Dean 0. Smith, and Paul S. G. Stein for helpful comments on earlier versions of the manuscript. We also thank Paul Stein for valuable discussions during the course of this work and Cheryl Adams for reconstruction of the hamster skeleton and preparation of the figures. This research was supported by NSF Grant BNS-8311517 and NIH Grant NS 14428 to K.K.
AMSTRONG, D. M. 1986. Supraspinal contributions tion and control of locomotion in the cat. Prog.
14.
to the initiaNeurobiol. 26:
15’ re.
3.
D. in the rat. BERNSTEIN, corticospinal postnatal
H. 1934. The results of unilateral pyramidal J. Comp. Neural. 60: 45-56. D. R., AND D. J. STELZNER. 1983. Plasticity tract following midthoracic spinal injury rat. J. Comp. Neural. 221: 382-400.
section of the in the
4. BREGMAN,
B. S., E. KUNKEL-BAGDEN, M. MCATEE, AND A. O’NEILL. 1989. Extension of the critical period for developmental plasticity of the corticospinal pathway. J. Comp. Neural.
282:355-370.
76:105-122.
20.
CHEEMA, S. S., A. RUSTIONI, AND B. L. WHITSEL. 1984. Light and electron microscopic evidence for a direct corticospinal projection to superficial laminae of the dorsal horn in cats and monkeys. J. Comp. Neurol. 225: 276-290.
11.
COHEN, A. H., AND C. GANS. 1975. Muscle activity in rat locomotion: Movement analysis and electromyography of the flexors and extensors of the elbow. J. Morphol. 146: 177-196.
12.
COULTER, J. D., AND E. G. JONES. 1977. Differential distribution of corticospinal projections from individual cytoarchitectonic fields in the monkey. Brain Res. 129: 335-340.
13.
DONATELLE,
J. M.
197’7. Growth
of the corticospinal
tract
The dur-
23.
of CinRes.
G. E., R. M. REINKING, AND D. G. STUART. 1973. cat step cycle: Hind limb joint angles and muscle lengths ing unrestrained locomotion. J. Morphol. 141: l-42. GRUNER, J. A., J. ALTMAN, AND N. SPIVACK. 1980. Effects arrested cerebellar development on locomotion in the rat: ematographic and electromyographic analysis. Ezp. Bruin
40:361-373. 24. HALBERTSMA,
J. M. 1983. The stride cycle of the cat: The modelling of locomotion by computerized analysis of automatic recordings. Acta Physiol. Stand. (Suppl.) 521: l-75.
25. HICKS,
S. P., AND C. J. D’AMATO. 1975. Motor-sensory cortexcorticospinal system and developing locomotion and placing in rats. Am. J. Anat. 143: l-42.
26. JENKINS,
F. A. 1974. The movement of the shoulder late and aclaviculate mammals. J. Morphol. 144:
in clavicu-
71-84.
K., AND G. E. SCHNEIDER. 1975. Motor performance lowing unilateral pyramidal tract lesions in the hamster. Res. 100: 170-174.
study
K., AND T. REH. 1982. A light of regrowing pyramidal tract
folBruin
and electron microscopic fibers. J. Comp. Neurol.
211:265-275. 29. KEIFER, J., AND K. KALIL. pyramidal tract cord preparation
640. 30. KEIFER, 31.
J., hamsters Neurosci. KEIFER, J., pyramidal 1030.
1989. Modulation of spinal reflexes by stimulation in an in uitro brainstem-spinal from the hamster. Exp. Bruin Res. 78: 633-
AND K. KALIL. 1987. Development of locomotion in with neonatal lesions of the pyramidal tract. Sot. Abstr. 13: 1503. AND K. KALIL. 1985. Functional role of the hamster tract during locomotion. Sot. Neurosci. Abstr. 11:
32. KUANG, 33.
R. Z., AND K. KALIL. 1989. Sprouting of corticospinal fibers into denervated contralateral spinal cord reveals specificity of new axon arbors. Sot. Neurosci. Abstr. 15: 92. KUANG, R. Z., AND K. KALIL. 1990. Branching patterns of corticospinal axon arbors in the rodent. J. Comp. Neurol., in press.
34. LAWRENCE, and
of pyramidal lesions Neurobiol. Exp. 40:
22. GOSLOW,
387. 10.
in the
31: 69-80.
AND J. HALBERTSMA. 1980. The cat. I. Coordination within a hind108: 269-281.
GORSKA, T., AND E. SYEURSKA. 1980. Effects on forelimb movements in the cat. Acta 843-859.
28. KALIL,
CHAPMAN, C. E., AND M. WIESENDANGER. 1982. Recovery of function following unilateral lesions of the bulbar pyramid in the monkey. Electroencephnlogr. Clin. Neurophysiol. 53: 374-
on interneurons
GILMAN, S., L. A. MARCO, AND H. C. EBEL. 1971. Effects of medullary pyramidotomy in the monkey. II. Abnormalities of spindle responses. Brain 94: 515-530. GOLDBERGER, M. E. 1969. The extrapyramidal systems of the spinal cord. II. Results of combined pyramidal and extrapyramidal lesions in the macaque. J. Comp. Neurol. 135: l-26.
BREGMAN, B. S., AND M. E. GOLDBERGER. 1983. Infant lesion effect. III. Anatomical correlates of sparing and recovery of function after spinal cord damage in newborn and adult cats. Dev. Brain Res. 9: 137-154.
9.
tract effects J. Neurophysiol.
19.
7.
of
horn.
girdle
FORSSBERG, H., S. GRILLNER, locomotion of the low spinal limb. Acta Physiol. Scund.
27. KALIL,
tract
dorsal
of the shoulder
166: 279-292.
18.
B. S., AND M. E. GOLDBERGER. 1983. Infant lesion effect. II. Sparing and recovery of function after spinal cord damage in newborn and adult cats. Deu. Bruin Res. 9: 119-135.
CHAMBERS, W. W., AND C.-N. LIU. 1957. Cortico-spinal the cat. J. Comp. Neurol. 108: 23-56.
analysis of forelimb in the cat. J. Exp. Biol.
ENGLISH, A. W. 1978. Functional analysis of cats during locomotion. J. Morphol.
6. BREGMAN,
8.
A. W. 1978. An electromyographic during overground stepping
cat lumbar
5. BREGMAN,
B. S., AND M. E. GOLDBERGER. 1983. Infant lesion effect. I. Effect of neonatal spinal cord lesion on motor development in the kitten. Deu. Brain Res. 9: 103-117.
ENGLISH, muscles
17. FETZ, E. E. 1968. Pyramidal
273-361. 2. BARRON,
the development of placing reactions in the postnatal rat. J. Camp. Neural. 176:207-232. EIDELBERG, E., AND J. Yu. 1981. Effects of corticospinal lesions upon treadmill locomotion by cats. Exp. Bruin Res. 43: lOl-
103.
21.
REFERENCES 1.
KALIL
D. G., AND H. G. J. M. KUYPERS. 1968. The functional organization of the motor system in the monkey. I. The effects of bilateral pyramidal lesions. Bruin 91: 1-14.
EFFECTS
OF
PYRAMIDAL
35.
LEMON, R. 1988. The output map of the primate Trends Neurosci. 11: 501-506.
36.
LIDDELL, E. G. T., AND C. G. PHILLIPS. in the cat. Brain 67: 1-9.
37.
MERLINE, M., AND K. KALIL. 1990. Cell death of corticospinal neurons is induced by axotomy before but not after innervation of spinal targets. J. Comp. Neural. 296: 506-516. MESULAM, M. 1978. Tetramethyl benzidine for horseradish peroxidase neurochemistry: A non-carcinogenic blue reactionproduct with superior sensitivity for visualizing neural afferents and efferents. J. Histochem. Cytochem. 26: 106-117. PHILIPPSON, M. 1905. L’autonomie et la centralization dans le systeme nerveux des animaux. Trav. Lab. Physiol. Inst. Soluay. 7: l-208.
38.
39.
motor
TRACT
1944. Pyramidal
cortex.
LESIONS
ON
LOCOMOTION
105
40.
PRESTON, J. B., M. C. SHENDE, AND K. UEMURA. 1967. The motor cortex-pyramidal system: Pattern of facilitation andinhibition on motoneurons innervating limb musculature of the cat and baboon and their possible adaptive significance. In Neurophysiological Basis of Normal and Abnormal Activities (M. D. Yahr and D. P. Purpura, Eds.), pp. 61-72. Raven, New York.
41.
RALSTON, D. D., AND H. J. RALSTON. 1985. The terminations of corticospinal tract axons in the macaque monkey. J. Comp. Neurol. 242: 325-337.
42.
REH, T., AND K. KALIL. 1981. Development tract in the hamster. I. A light microscopic Neural. 200: 55-67.
43.
REH, T., AND K. KALIL. 1982. Functional midal tract fibers. J. Comp. Neural.
section
of the pyramidal study. J. Camp.
role of regrowing
211: 176-283.
pyra-