Initiation of locomotion from the mesencephalic locomotor region: Effects of selective brainstem lesions

Brain Research, 328 (1985) 121-128 Elsevier

121

BRE 10555

Initiation of Locomotion from the Mesencephalic Locomotor Region: Effects of Selective Brainstem Lesions R. M. JELL, C. ELLIOTT and L. M. JORDAN

Departmentof Physiology, The Universityof Manitoba, Facultyof Medicine, Winnipeg, Manitoba, R3E OW3 (Canada) (Accepted May 29th. 1984)

Key words: locomotion - - brainstem - - Deiters' nucleus--ventral tegmental area

The effects of selected brainstem lesions on controlled treadmill locomotion produced by stimulation of the mesencephalic locomotor region (MLR) in postmamillary cats were determined in these experiments. The importance for the initiation of locomotion of projections from the MLR to rostral brainstem structures, described in a preceding paper28. were examined by selective lesioning or by adjusting the level of the decerebration. The role played by the lateral vestibulospinal tract (LVST) in the initiation of locomotion was examined by lesioning Deiters" nucleus bilaterally. Contrary to previous claims 26, the results of the present experiments show that areas of the brainstem rostral to the MLR are not required for the initiation of locomotion by MLR stimulation. This finding eliminates the ventral tegmental area of Tsai and the substantia nigra, both implicated in the initiation of locomotion, as required participants in MLR stimulated locomotion. Bilateral Deiters" nucleus (DN) lesions did not significantly affect the initiation of locomotion from the MLR, nor did such lesions alter in a systematic fashion the amplitude or timing of EMG activity in flexor or extensor muscles of the hindlimb during MLR evoked walking. Joint angle changes during the locomotor cycle were also essentially unaltered by DN lesions. The significance of these findings regarding the brainstem structures which must be involved in the initiation of locomotion are discussed.

INTRODUCTION It was d e m o n s t r a t e d in a previous paper27 that controlled treadmill locomotion p r o d u c e d in decerebrate cats by stimulation of the mesencephalic locom o t o r region ( M L R ) is abolished by lesions of the ventrolateral funiculus ( V L F ) of the spinal cord. Abolition of locomotion in awake, otherwise intact cats and monkeys by V L F lesions has also been observedJ-5, and there is evidence that sparing of locom o t o r capability in humans with spinal cord injury requires that some portion of the ventral spinal cord be intact 3. A l t h o u g h these studies establish the spinal trajectory for a necessary pathway (or pathways) for the initiation of locomotion, they do not distinguish among the many fiber pathways traversing the V L F in terms of their relative importance for the initiation of locomotion. The projections of cells in the M L R have been described in recent studies using autoradi-

ographic tracing techniques 7.2s. Cells within the M L R do not a p p e a r to send a significant projection to the spinal cord directly, but they do terminate on the cells of certain brainstem areas known to project to the spinal cord through the V L F 2s. It is possible that interruption of these pathways accounts for the effect of V L F lesions on the initiation of locomotion. Destruction of any particular pathway in the V L F could conceivably block M L R stimulated locomotion through a variety of mechanisms, including interruption of an O N / O F F 'command" pathway relayed from the M L R via a lower brainstem center. Issues specifically addressed in this study include: (1) the possibility that the ascending projections from the M L R 2s and not the descending ones constitute the significant projections for the initiation of locomotion; and (2) the possibility that the V L F lesion interrupts some descending pathway with tonic effects on spinal neurons. Evidence supporting the former possibility is

Correspondence: L. M. Jordan, Department of Physiology, The University of Manitoba. Faculty of Medicine. Winnipeg. Manitoba. R3E 0W3, Canada. 0006-8993/85/$03.30 © 1985 Elsevier Science Publishers B.V.

122 provided by the observation that MLR stimulation is without effect in animals with midcollicular decerebrations 26, in which the MLR is intact but its rostral projections are largely eliminated. One of the rostral target sites for MLR projections is the ventral tegmental area of Tsai 28, which is known to project to cells of origin of fibers projecting to the spinal cord 2,3l, some of which travel in the VLF 12. Certain stimuli applied to the VTA can produce locomotor activity 15. The suggestion that VLF lesions may interrupt a descending pathway having a tonic influence on the neurons involved in the control of locomotion is supported by the finding that levels of extensor tone can determine whether MLR stimulation is effective in inducing locomotion or not 16. A candidate pathway for such an effect is the lateral vestibulospinal tract (LVST), which travels in the VLF 12.13.Ls, and causes enhanced extensor activity22 or resetting of the locomotor rhythm 24 when its cells of origin in Deiters' nucleus (DN) are stimulated. Orlovsky 23 reported that unilateral lesions of DN markedly reduced or abolished ipsilateral extensor activity during controlled treadmill locomotion. Yu and Eidelberg 33 reported a marked reduction in the extensor components of stepping in freely moving animals walking on a treadmill after DN lesions. We examined the importance of projections from the MLR to rostral brainstem structures, described in a preceding paper 28, by selective lesioning or by adjusting the level of the decerebration. We also sought to determine whether the effects of lesions in the ventrolateral quadrant of the spinal cord on the initiation of locomotion from the MLR could be due to the interruption of the LVST. A preliminary report of some of the data has been presented ~1. MATERIALS AND METHODS All experiments were performed on adult cats weighing between 2.0 and 5.0 kg. Each animal was anesthetized by a mixture of nitrous oxide and halothane administered through an endotracheal tube. Both common carotid arteries were ligated and one was also cannulated for the purpose of monitoring blood pressure. One external jugular vein was cannulated to allow for the administration of fluids. Intramuscular EMG electrodes were placed bilaterally in some or all of the following muscles for the purpose

of recording muscle activity during locomotion: iliopsoas (IP), adductor magnus (AM), posterior biceps (PB), vastus lateralis (VL), tibialis anterior (TA) and medial gastrocnemius (MG). The head of the animal was mounted in a stereotaxic device suspended over a treadmill, and the hindquarters of the animal were suspended using a foam rubber hammock positioned just anterior to the hindlimbs. The lateral aspect of the left hindlimb and pelvic region was shaved in some cats, and the skin was marked with colored dots approximately 0.5 mm in diameter at points over the iliac crest, the ischial tuberosity, the head of the femur, the lateral maleolus of the fibula and the peroneal tubercle of the calcaneus. These dots were used to establish the changes in hip, knee and ankle joint angles during locomotion by kinematic analysis from filmed records. Because the skin over the knee joint tended to move relative to the underlying joint, the location of that point was computed from measurements of femur and fibula length. Postmammillary preparations were produced by performing a decerebration with a spatula along a plane extending from the rostral border of the superior colliculus dorsally to the caudal border of the mammillary bodies ventrally (line B, Fig. 1). These preparations were used in those experiments where subsequent lesions were made, or where a second decerebration at a more caudal level of the brainstem was required. In 3 additional animals, a midcollicular transection (line C, Fig. 1) was performed at the beginning of the experiment by transecting the brainstem along a plane extending from the intercollicular line dorsally to the rostral border of the pons ventrally. Locomotion was evoked by electrical stimulation with a stainless steel monopolar electrode having a diameter of 0.1 mm at the tip and insulation to within 0.2 mm of the tip. The stimulus consisted of a train of constant current pulses 0.5 ms in duration at a rate of 30 Hz applied throughout the locomotion trial period. Stimulus strengths were in the range of 25-200 uA. The MLR was localized along a vertical electrode tract situated 4 mm lateral to the midline and 1 mm posterior to the junction between the inferior and superior colliculi. The ultimate selection of the site of lowest threshold for initiation of locomotion sometimes required placement of the stimulating electrode in closely adjacent vertical tracts. Control

123 locomotor trials were recorded prior to subsequent brainstem lesioning procedures. Lesions of the ventral tegmental area (VTA) were made bilaterally in 4 animals using radio frequency electrical current passed between adjacent pairs of stainless steel wires insulated except for 2 mm at the tip. The size of the lesion was adjusted to include the entire VTA. In 4 cats, an intercollicular decerebration was performed subsequent to a previous postmammillary transection and control trials. Another 6 cats were decerebrated at the midcollicular level at the beginning of the experiment and then subjected to MLR stimulation. Bilateral Deiters' nucleus (DN) lesions were produced in 10 animals after postmammillary decerebration. Control locomotor trials were attempted in these animals after a sham lesioning procedure, which consisted of exposure of the cerebellum and elevation of the vermis to reveal the cerebellar peduncles and allow access to the lesioning sites. When the control trials were completed, the animal was reanesthetized, the vermis was again elevated and DN was lesioned bilaterally by coagulation with the end of a hot probe6 made of 1 mm diameter steel wire, bent at an angle of 135° at a distance of 5 mm from the end. Disappearance of decerebrate rigidity was observed bilaterally in all cases following recovery from the anesthetic. Kinematic analysis of locomotor activity was carried out through the use of filmed records made with a Super-8 movie camera (24 or 36 frames/s). Each frame was projected onto a digitizing tablet under the control of a Tektronix 4051 Desktop Computer. Coordinates of each of the marks on the hindlimb were automatically sensed by tablet pen position, then stored in digital form on magnetic tape for later plotting of joint angles and stick figure representation of hindlimb position for each portion of the step cycle. EMG recordings on magnetic tape were replayed with analog full wave rectification and integration (20 ms time constant) of each channel. All integrated channels were digitized simultaneously at 100 Hz per channel using a microcomputer. Analysis of the digitized EMG records consisted of normalization of each channel with respect to a selected channel, usually MG. This channel was used to determine the corn-

mencement of the step cycle for each step. Variability in step cycle length was taken into account by assuming that all channels had the same duration as the longest cycle in the reference channel, then filling in any missing data points for the other steps by interpolation. This allowed averaging of all step cycles for each muscle and the computation of a mean normalized EMG curve and the standard deviation for each channel. The points of onset and termination of EMG activity were then easily determined from the normalized curves using either a computer detection method or manually by taking as the point of onset the beginning of a deviation in the waveform which produced an increase in the amplitude of more than 2 S.D. above the baseline level of activity. The point of termination of activity was determined in a similar manner. At the end of each experiment the brainstem was removed, fixed in formalin, and subsequently sectioned and stained using standard techniques for verification of lesion sites by microscopic examination. Drawings of the extent of the lesion and the outline of the tissue sections were made for identification of structures in the area of the lesion (see Fig. 3). The exact level of decerebration was determined by inspection of the rostral end of the brainstem after its removal from the animal. RESULTS Lesions of areas rostral to the M L R The VTA was lesioned bilaterally in 4 animals subsequent to decerebration and localization of the most effective site for MLR stimulation. Locomotion could be induced by stimulation of the same MLR site within 1 h of the lesion in all cases. The threshold for the induction of locomotion (50-100/~A) was unchanged after the lesion was made. Locomotion was coordinated and not obviously different from that seen in the same animals prior to the lesion. Histological verification of the lesion sites in these animals revealed that the lesions included the VTA as well as some surrounding tissue in each case. The absence of effect of lesioning the major projection from the MLR to the area beneath the superior colliculus prompted a re-evaluation of the claim that this area must be intact in order for MLR stimulation to be effective 26. Locomotion could be induced

124

5

,0 LAT. ,

20

1

I5

l

;

10

I 0

5

4.1

I A 5

|l 0

Fig. 1. Parasagittal brainstem section illustrating levels of transection. The plane represented by the line labeled A corresponds to the plane of transection for premammiIlary preparations, B, for postmammillary preparations and C, for intercollicular preparations. Abbreviations: BC, brachium conjunctivum; CNF, cuneiform nucleus (MLR); IC, inferior colliculus; MB, mammillary bodies; OT, optic tract; PG, pontine gray matter; SUB, subthalamic nucleus; NIII, third cranial nerve.

after the midcollicular transection of the brainstem. MLR stimulation continued to be effective for the initiation of locomotion in this case, and several spontaneous steps occurred after the treadmill was switched on, prior to the commencement of MLR stimulation. The quality of locomotion in animals with a midcollicular transection was not obviously different from that in animals with postmammillary decerebrations. The thresholds for MLR evoked locomotion after intercollicular decerebration were within the range of 50-150 ~A. A similar range of effective strengths has been observed in other studies using the same type of electrodes and stimulation p r o c e d u r e s 27,29.3°. In those cases where a second transection was produced, the threshold did not change significantly after the additional lesion. Deiters' nucleus lesions T h e e f f e c t of b i l a t e r a l d e s t r u c t i o n o f D N o n M L R

by MLR stimulation in 7 of the 10 animals subjected to intercollicular decerebration. Fig. 2 illustrates the EMG activity in bilateral extensor muscles of the hindlimb during MLR evoked locomotion before and

e v o k e d l o c o m o t i o n was e v a l u a t e d in 10 cats in w h i c h t h e l e s i o n was f o u n d t o i n c l u d e all o r m o s t o f t h e n u cleus o n e a c h side. T h e e x t e n t of a typical l e s i o n , rec o n s t r u c t e d f r o m serial s e c t i o n s o f t h e b r a i n s t e m , is

A ~il

~"-

ilL.

Jl=

=,=- 1 ~ .

.~lk

dllLJUl,

=i~

i~

L

.lUll

JlIIL,II,

.~

.~i,

aulL

L MG = , = ,,~IL - . ~

=dl

,i,

,,.,~ ~

R MG i off

B L MG

'

'

1 sec.

l

Oil

l

off

Fig. 2. EMG activity recorded in the left and right MG muscles during locomotion produced by MLR stimulation after an initial postmammillary decerebration (A) and after a subsequent midcollicular decerebration (B). The same MLR stimulus site and stimulus parameters (30 Hz, 0.5 ms, 75 ~A) were used in both cases. MLR stimulation was begun a few seconds prior to the beginning of the traces in A; the points of onset and termination of MLR stimulation are indicated by arrows elsewhere in the figure.

125 Pf~

"~

CB VSN~-

RS [~

VLD (~ -5

-5

I0 .

•

.

|

i

i

5

|

|

i

i

i

0

i

i

i

i

i

i

i

,

i

I0 i

,

I

i

i

i

i

i

i

I

i

5

5

P

a.s

|

0

i

i

i

i

I

i

i

i

i

i

5

L----.....,

O

J

-5

-I0 i

i

i

i

•

5

i

I

I

i

i

0

i

i

i

i

i

i

i

•

•

5

Fig. 3. Drawings of brainstem cross sections showing the lesion sites (shaded areas) reconstructed from a typical experiment in which bilateral destruction of DN was performed. illustrated in Fig. 3. In all cases, extensor rigidity produced by the decerebration was absent after recovery from the anesthetic. However, during the course of recovery and subsequent to M L R stimulation extensor tone increased in 5 animals to levels sufficient for weight support during the intervals between locomotor trials. This was established by lowering the animal onto the surface of the treadmill until the hindlimbs began to bear the weight of the hindquarters. Extensor activity was always sufficient for weight support during MLR induced locomotion, as well. Five of the animals subjected to DN lesions could be made to walk with MLR stimulation within a mean of 130 min after the lesion was completed. Thresholds for evoking locomotion were in the range of 25-200 # A before and after the DN lesions. No significant difference was found between the pre-lesion thresholds and those required after recovery. The other 5 animals never fully recovered from the lesion, and their condition, as judged by blood pressure and respiration indices, deteriorated steadily until

the experiment was terminated. These same animals never recovered any significant extensor tone. MLR induced locomotion was evaluated after a sham lesioning procedure and after DN lesions using kinematic and normalized E M G measures. Fig. 4 illustrates the hip, knee and ankle joint angles throughout the step cycle before and after DN lesioning. There were no qualitative differences in the movements about the joints in the hindlimb in this animal or in 5 other cats which could be made to walk with MLR stimulation. Evaluation of the stepping produced by MLR stimulation after DN lesions using normalized E M G is illustrated in Fig. 5. Extensor (MG) and flexor (TA) activity patterns are illustrated for two animals before and after the DN lesions. Extensor activity could either increase (Fig. 5A) or decrease (Fig. 5C) after the lesion, with no change in flexor muscle activity. Overall, the DN lesions did not produce any consistent alteration in the levels of E M G activity during MLR induced locomotion. There was one animal in which the amplitude of the normalized extensor

126

A

B

~4~

14e

Ohima

r'lknee

r-lk~ee

~ankle

/~Qnkle 40

start

SIEP

e~d

4e

start

STEP

end

Fig. 4. Kinematicanalysis of the step cyclebefore (A) and after (B) bilateral DN lesions. The stick figure progression above each chart shows successive hindlimb joint angles corresponding in time to the plotted joint angles on the graphs. The vertical coordinate of the hip joint was kept constant during the plot of the stick figures. One step cycle is plotted from left to right starting with the commencement of the swingphase. EMG was judged to be significantly increased after the lesion based upon a difference of more than 2 S.D. between the area under the mean curve for MG in the control condition and that of the same muscle after bilateral DN lesions. The MG EMG amplitude was increased by more than 1 S.D. in a second preparation. In two animals the amplitude of MG activity was significantly decreased after the DN lesions, with the normalized EMG during the control condition being more than 1 S.D. greater than the EMG from the same muscle after the lesions. In one additional preparation there was no significant change in the MG EMG after the lesions. Flexor (TA) EMG amplitude never varied more than 1 S.D. Furthermore, the times of onset and termination of EMG activity were not significantly altered by the lesions. Thus, DN lesions neither blocked the pathways by which MLR stimulation can initiate locomotion, nor produced any predictable alterations of intensity and timing of muscle activity. DISCUSSION The results of these experiments reveal that neither the VTA nor other structures in the rostral midbrain and known to receive projections from the MLR are essential for the initiation of locomotion.

This suggests that certain of the MLR's descending projections are important for the relay of the locomotion initiation signal to the spinal cord. Nevertheless, it is possible that a portion of the periaqueductal gray matter in the vicinity of the inferior colliculus, which remains intact in the intercollicular preparation, may be of importance in the initiation of locomotion by MLR stimulation. It is also evident that previous claims that locomotion cannot be produced by MLR stimulation in animals with an intercollicular decerebration 26 are incorrect. The substantia nigra, which has been stimulated to produce locomotion8 in postmammillary animals, likewise appears to be non-essential for the initiation of locomotion from the MLR on the basis of the present results. These experiments also reveal that interruption of the LVST cannot account for the fact that VLF lesions abolish locomotion evoked from the MLR, because bilateral DN lesions have no lasting effect on MLR evoked locomotion. Clearly this evidence supports the idea that the initiation of locomotion depends more upon the integrity of pathways descending from the sites of termination of MLR cells in the ponto-medullary reticular formation. We have shown25 that cooling the sites of termination of fibers from the MLR in the lower brainstem 2s reversibly blocks locomotion produced by MLR stimulation.

127

A

CONTROL L MG

LESIONED

Anatomical studies of these regions have shown that

L MG

the gigantocellular and magnocellular tegmental fields both receive connections from the M L R 7,2s and send p r o m i n e n t projections to the spinal cord which descend in the V L F 1,12A3,31,32. Eiectrophysiological [,

,.,~'L~ R TR

R

evidence 19-2~ confirms that M L R stimulation acti-

TR

vates reticulospinal cells in the magnocellular and gigantocellular reticular formation monosynaptically. Furthermore, locomotor activity has been produced in one study by stimulation of these areas~7. A further issue addressed by this study is the possiB

630700660650640690560660700

10660620?30

870

L

g T

C

CONTROL L MG

L

TR

LESIONEI] L MG

L

TR

ble role of DN outflow on the m a i n t e n a n c e of the normal locomotor rhythm produced in response to M L R stimulation. Orlovsky 22 showed that stimulation of DN, which has potent effects on spinal motoneurons 9:0,14,32, could enhance E M G activity in hindlimb extensor muscles, while Russell and Zajac 24 found that DN stimulation could strongly influence the timing and pattern of the locomotor rhythm during L - D O P A induced fictive locomotion. Yu and Eidelberg 33 reported a reduction in extensor activity due to DN lesions in otherwise intact chronic cats

Fig. 5. Effects of bilateral DN lesions on ankle flexor and extensor muscle EMG activity during locomotion. An experiment in which MG EMG was enhanced after the lesion is shown in A, Heavy lines represent the mean EMG activity during a 10 s period of walking, and the dashed lines represent the S.D. The traces in B represent the rectified EMGs of left MG and right TA from which the averaged records in A (lesioned) were taken. The vertical bars indicate the beginning of each step, determined from the beginning of EMG activity in left MG. The numbers above the left MG trace indicate the durations of the steps in ms. TA activity was essentially unchanged in both cases. The procedure for EMG averaging is provided in detail in the text.

walking on a treadmill. Our results show that the extensor E M G activity is not diminished during treadmill walking in postmammillary cats. This apparent discrepancy between effects obtained on otherwise intact animals and our postmammillary ones could be due to the fact that our animals were suspended by the stereotaxic apparatus and the abdominal hammock, resulting in removal of the r e q u i r e m e n t for lateral stability and elimination of head accelerations. Nevertheless, our results suggest that D N is not a major contributor to the extensor activity accompanying the u n p e r t u r b e d locomotor rhythm produced by MLR stimulation.

REFERENCES 1 Basbaum, A. I., Clanton, C. H. and Fields, H. L., Three bulbospinal pathways from the rostral medulla of the cat: an autoradiographic study of pain modulating systems, J. comp. Neurol., 178 (1978) 209-224. 2 Chi, C., Bandler, R. and Flynn, J. P., Neuroanatomic projections related to biting attack elicited from ventral midbrain in cats, Brain Behav. Evol., 13 (1976) 91-110. 3 Eidelberg, E., Consequences of spinal cord lesions upon motor function, with special reference to locomotor activity, Progr. Neurobiol., 17 (1981) 185-202. 4 Eidelberg, E., Story, J. L., Walden, J. G. and Meyer, B. L., Anatomical correlates of return of locomotor function

5 6 7

8

after partial spinal cord lesions in cats, Exp. Brain Res., 42 (1) (1981) 81-88. Eidelberg, E., Walden, J. G. and Nguyen, L. H., Locomotor control in macaque monkeys, Brain, 104 (1981) 647-663. Fulton, J. F., Liddell, E. G. T. and Rioch, D. Mc. K., The influence of unilateral destruction of the vestibular nuclei upon posture and the knee jerk, Brain, 43 (1930) 327-343. Garcia-Rill, E., Skinner, R. D., Gilmore, S. A, and Owings, R., Connections of the mesencephalic locomotor region (MLR). II. Afferents and efferents, Brain Res. Bull., 10 (1983) 63-71. Garcia-Rill, E., Skinner, R. D., Jackson, M. B. and Smith, M. M., Connections of the mesencephalic locomotor region

128 (MLR). I. Substantia nigra afferents, Brain Res. Bull., 10 (1983) 57-62. 9 Grillner, S., Hongo, T. and Lund, S., The vestibulospinal tract: effects on alpha motoneurones in the lumbosacral spinal cord in the cat, Exp. Brain Res, 10 (1970) 94-120. 10 Grillner, S., Hongo, T. and Lund, S., Convergent effects on alpha motoneurons from the vestibulospinal tract and a pathway descending in the medial longitudinal fasciculus, Exp. Brain Res., 12 (1971) 457-479. 11 Jell, R. M., Jordan, L. M. and Ireland, D. J., Evoked locomotion after destruction of Deiters' nucleus, Soc. Neurosci. Abstr., 7 (1981) 687. 12 Kuypers, H. G. J. M. and Maisky, V. A., Funicular trajectories of descending brainstem pathways in the cat, Brain Research, 136 (1977) 159-165. 13 Kuypers, H. G. J. M. and Maiskey, V. A., Retrograde axonal transport of HRP from spinal cord to brain stem cell groups in the cat, Neurosci. Leg., 1 (1975) 9-14. 14 Lund, S. and Pompeiano, O., Monosynaptic excitation of alpha motoneurones from supraspinal structures in the cat, ActaphysioL scand., 73 (1968) 1-21. 15 Mogenson, G. J., Wu, M. and Manchanda, S. K., Locomotor activity initiated by microinfusions picrotoxin into the ventral tegmental area, Brain Research, 161 (1979) 311-319. 16 Mori, S., Kawahara, K., Sakamoto, T., Aoki, M. and Tomiyama, T., Setting and resetting of level of postural mus-cle tone in decerebrate cat by stimulation of brain stem, J. NeurophysioL, 48 (1982) 737-748. 17 Mori, S., Nishimura, H., Kurakami, C., Yamamura, T. and Aoki, M., Controlled locomotion in the mesencephalic cat: distribution of facilitatory and inhibitory regions within pontine tegmentum, J. Neurophysiol., 41 (1978) 1580-1591. 18 Nyberg-Hansen, R. and Mascitti, T. A., Sites and mode of termination of the vestibulospinal tract in the cat. An experimental study with silver impregnation methods, J. comp. NeuroL, 122 (1964) 369-388. 19 Orlovsky, G. N., Connexions of the reticulo-spinal neurones with the 'locomotor sections' of the brain stem, Biofizika, 15 (1970) 171-177. 20 Orlovsky, G. N., Electrical activity in the brainstem and descending pathways in guided locomotion, Sechenov physiol. J. U.S.S.R., 55 (1969) 437-444.

21 Orlovsky, G. N., Work of reticulospinal neurones during locomotion, Biofizika, 15 (1970) 728-737. 22 Orlovsky, G. N., The effect of different descending systems on flexor and extensor activity during locomotion, Brain Research, 40 (1972) 359-371. 23 Orlovsky, G. N., Activity of vestibuiospinal neurons during locomotion, Brain Research, 46 (1972) 85-98. 24 Russell, D. F. and Zajac, F. E., Effects of stimulating Deiters' nucleus and medial longitudinal fasciculus on the timing of the fictive locomotor rhythm induced in cats by DOPA, Brain Research, 177 (1979) 588-592. 25 Shefchyk, S. J., Jell, R. M. and Jordan, L. M., Reversible cooling of the brainstem reveals areas required for mesencephalic locomotor region evoked treadmill locomotion, Exp. Brain Res., in press. 26 Shik, M. L., Severin, F. V. and Orlovsky, G. N., Structures of the brain stem responsible for evoked locomotion, Fiziol. Zh. (Leningr. ), 12 (1967) 660-668. 27 Steeves, J. D. and Jordan, L. M., Localization of a descending pathway in the spinal cord which is necessary for controlled treadmill locomotion, Neurosci. Lett., 20 (1980) 283-288. 28 Steeves, J. D. and Jordan, L. M., Autoradiographic demonstration of the projections from the mesencephalic locomotor region, Brain Research, 307 (1984) 263-276. 29 Steeves, J. D., Jordan, L. M. and Lake, N., The close proximity of catecholamine-containing cells to the 'mesencephalic locomotor region' (MLR), Brain Research, 100 (1975) 663-670. 30 Steeves, J. D., Jordan, L. M., Schmidt, B. and Skovgaard, B. J., Effect of norepinephrine and 5-hydroxytryptamine depletion on locomotion in the mesencephalic cat, Brain Research, 185 (1980) 349-362. 31 Swanson, L. W., The projections of the ventral tegmental area and adjacent regions: a combined fluorescent retrograde tracer and immunofluorescence study in the rat, Brain Res. Bull., 9 (1982) 321-353. 32 Wilson, V. J. and Yoshida, M., Comparison of the effects of stimulation of Deiters' nucleus and medial longitudinal fasciculus on neck, forelimb, and hindlimb motoneurons, J. NeurophysioL, 32 (1969) 743-758. 33 Yu, J. and Eidelberg, E., Effects of vestibulospinal lesions upon locomotor function in cats, Brain Research, 220 (1981) 179-183.