Brain Research, 172 (1979) 197-208 © Elsevier/North-Holland Biomedical Press
197
Research Reports P E R I P H E R A L I N P U T PATHWAYS P R O J E C T I N G TO THE MOTOR CORTEX IN T H E CAT
I-I. ASANUMA, K. D. LARSEN and P. ZARZECKI* The Rockefeller University, New York, N. Y. 10021 (U.S.A.)
(Accepted December 24th, 1978)
SUMMARY The possibility that the motor cortex receives peripheral input directly from the thalamus was examined using the evoked potential method and the following results were obtained. Potentials in the motor cortex evoked by stimulation of superficial radial (SR) or group II deep radial (DR) nerve were neither abolished nor delayed by ablation of the sensory cortex. Potentials in the motor cortex evoked by stimulation of group II D R nerve were most severely reduced by interruption of the spinocervical tract. Potentials evoked by stimulation of SR nerve were more severely reduced in the sensory cortex than in the motor cortex by section of the dorsal funiculus or cooling of the cuneate nucleus. The size of evoked potentials in the motor cortex increased rapidly when stimulus intensity to D R nerve exceeded the threshold to group II fibers. The results suggest that some inputs from the SR and group II D R nerves reach the motor cortex without a relay through the sensory cortex. INTRODUCTION Cells of the motor cortex receive topographically organized peripheral inputs 7, 11, but the pathway through which these inputs travel has not been fully defined. The motor cortex receives inputs from several cortical areas and subcortical nuclei. Cells of the thalamic nucleus ventralis lateralis which constitute the major input to motor cortex are not responsible for the somesthetic inputs because they do not receive precise information from the periphery3,21, 22. The sensory cortex projects to motor * Present address: Department of Physiology, Queen's University, Kingston, Ontario, Canada K7L 3N6.
198 cortex 1~ and is a possible source of the somesthetic inputs. This possibility was further emphasized by Wiesendanger 31, on the basis of the difference of lateneies for unit discharges in the motor and the sensory cortices. However, other evidence suggested that the sensory cortex cannot be the only source: peripherally evoked potentials in the monkey motor cortex do not disappear after ablation 2° or cooling 27 of the sensory cortex, and the latencies of the evoked potentials are not longer in the motor than in the sensory cortex 27. If there is a direct input from the periphery to the motor cortex, it may constitute the simplest cortical reflex pathway which could possibly be the basis of various cortical reflexes. Because of the potential importance of such a reflex, we wished to determine if the sensory inputs to the motor cortex arrive through the sensory cortex or if they arrive directly from the thalamus independent of the sensory cortex. This has been studied in several ways and is presented in this and the following two papers 4,19 Cats were used for the experiments since the relationships between afferent input and efferent outflow of the feline motor cortex 6 are similar to those in monkeys 27 and man 12, and because they are readily available. In the first series of experiments, the question of whether the evoked potentials in the motor cortex are dependent on the peripheral inputs which come through sensory cortex is examined by removing the sensory cortex or making selected spinal tract sections. It will be shown that some somesthetic inputs to the motor cortex do not come through the sensory cortex. These direct inputs to the motor cortex travel largely in the spinocervical tract whereas the inputs to the sensory cortex travel both in the spinocervical tract and the dorsal fasciculus. The thalamic nuclei responsible for the relay of the sensory input to motor cortex will be considered in the following two papers4, ~9. METHODS Experiments were performed on 14 cats of either sex weighing between 2.5 and 3.5 kg, anesthetized with pentobarbital sodium (Nembutal, 35 mg/kg). The pericruciate and coronal sulci on both sides were exposed by opening the skull and were covered by a mixture of mineral oil and vaseline. The superficial and deep radial nerves of both sides were dissected and mounted on bipolar stimulating electrodes in mineral oil pools made of skin flaps. The cervical cord was exposed from C-1 to C-4 by laminectomy and the dura was opened to provide access to dorsal or lateral funiculus or both during the experiments. In addition, in 6 cats the foramen magnum was enlarged and the caudal cerebellar vermis was partly removed to allow placement of a cooling probe on the surface of the caudal medulla over the cuneate nucleus.
Stimulating and recording Bipolar Ag-AgC1 electrodes were used to stimulate the deep radial (DR) and the superficial radial (SR) nerves. Single pulse stimuli were delivered at 2-3 sec intervals and the evoked potentials were recorded from the surface of the cerebral cortex by silver ball electrodes placed around the cruciate sulcus (motor cortex), coronal sulcus
199 (sensory cortex) and post-cruciate dimple (area 3a of Hassler and Muhs-Clement14). Potentials were recorded simultaneously from two cortical sites, and were averaged on-line (Fabri-Tek, model 1062) and displayed on an oscilloscope.
Cooling the euneate nucleus In 6 experiments a cooling probe with a ventral surface of 3 x 10 m m was positioned so that it was lightly touching the surface of the caudal medulla over the nuclei gracilis and cuneatus. While recording from the cortex, saline cooled to 0 °C was pumped through the probe.
Histological examination Whenever cortex was removed or the dorsal and/or lateral funiculi were cut, the cortex or cervical cord was removed at the end of the experiment and fixed in 10~o formalin. Frozen sections were made and stained by Kliiver and Barrera's method 16 to reconstruct the cut or lesioned area. RESULTS
Characteristics of cortical-evoked potentials The distribution of evoked potentials elicited by stimulation of superficial radial (SR) and deep radial (DR) nerves was examined first. Although this has been studied before, as summarized by Woolsey 82, the recent development of an on-line averaging computer made it possible to determine the distribution with greater accuracy. Two cats were used for this purpose and provided similar results. The evoked potentials were elicited by stimulation with two times the threshold strength which was supramaximal for cortical-evoked potentials. The potentials were recorded from 120 spots around the cruciate and coronal sulci, forming a grid of 1.0 m m squares. At each point, 32 evoked potentials were averaged and the voltage differences between the peaks of the initial positive phase and the following negative phase were measured as shown by the inset diagram in Fig. 1. Isopotential maps were constructed for both the
DR f F
Fig. 1. Distribution of evoked potentials elicited by stimulation of superficial radial (SR) and deep radial (DR) nerves. Supramaximal stimuli (3 x threshold stimuli) were used. The voltage difference between the positive and the negative waves were measured as shown by the inset diagram and equipotential points were connected at 50 #V steps, the outermost lines being 50/~V. Further details in the text.
200
Control
S Cx removed
I
__J ,oo,, 5msec
5reset
2rr
AS
,CrS
Fig. 2. SR- and DR-evoked potentials in the motor cortex before and after removal of the sensory cortex. Arrows on the evoked potentials indicate the arrival of the afferent volleys. The brain photograph and the sketch of histological slide show the extent of the aspiration. AS, ansate sulcus; CrS, cruciate sulcus; CoS, coronal sulcus. SR- and DR- (Fig. 1A, B) evoked potentials. In agreement with a study carried out under similar conditions 2z, SR stimulation elicited large evoked potentials at the lateral tip of the cruciate sulcus, lateral to the coronal sulcus, and at the post-cruciate dimple. D R stimulation elicited large potentials at the post-cruciate dimple and lateral to the cruciate sulcus but only small potentials lateral to the coronal sulcus as reported previously25, z0. In the next trial, the sensory cortex was removed as widely as possible in two cats to determine its possible contribution to the evoked potentials in the motor cortex. The results of both experiments were similar and one of them is illustrated in Fig. 2. The area of sensory cortex which was removed (inset photograph, Fig. 2) included the post-cruciate dimple (area 3a), coronal gyrus (area 3b, 1) and suprasylvian gyrus (area
201
A
B DR sttm
SR sl~m
o
100
o
"
100
~80
~
~oo
~ oo.
i
~.
,- 40 20 ~
8O
40.
•
13~T
o.--
-
"
_ _
!
/ "-,/I
20.
/...............~" 1.0
[
112
" 1.4 " 1.6 Stimulus Strength x T
---
~T
r-.,._ ~-
",
"
118
10
11
1.2 13 14 St~nulus Strength x
1.5 T
!~/ / ~ .
]%
"
I
v
:,,;.'"
Fig. 3. Relationships between stimulus intensity and the amplitude of evoked potentials for DR and SR nerves. A, response curves for DR stimulation. Specimen-evoked potentials are shown underneath the graph. Calibration pulses on the specimen records are 50 ttV, 5 msec. An arrow on the specimen (1.7 × T) indicates arrival of the incoming volley. B : curves for SR stimulation. Further details in the text. 2) although the depths of cortex around the coronal sulcus were spared. Recordings were made 3 h after removal of the sensory cortex. Both SR- and DR-evoked potentials became much smaller (about 1/3 of the control) after the removal. Furthermore, the shape of the SR-evoked potentials changed in both cases as shown in the upper traces. However, the latencies of both SR- and DR-evoked potentials were not altered by the lesion. It is known that the amplitude of the DR-evoked potential in area 3a is proportional to the amplitude of group I volleys in the D R nerve, and that an increase in D R stimulation beyond 1.6-2.0 times the threshold intensity for group I fibers does not increase the size of the DR-evoked potentials 2~. We examined the relationship between the intensity of the stimulus and the amplitude of DR-evoked potentials at the dimple (area 3a) and the m o t o r cortex. The experiments were carried out in 3 cats which provided similar results. The threshold strength for eliciting cortical-evoked potentials was taken as threshold strength for SR or D R nerve fibers as has been described previously 25,2s. The amplitude of DR-evoked potentials in area 3a increased rapidly as the stimulus intensity was increased from 1.0 to 1.4 times the threshold (Fig. 3A). In the m o t o r cortex, however, the amplitude started increasing rapidly at around 1.5 times the threshold. Since the recruitment of group II fibers starts at around 1.5 times the threshold strength for group I fibers is, the components of DR-evoked potentials recruited above 1.5 times the threshold are likely to be elicited by inputs
16
202
¢N
.;.:::,.
}
/
.. Ci o// / rn
~
.
i-:
.
gSUOdSe8
IOJ1UO~) %
¢0
,',|
1~
oY
!L
197
Research Reports P E R I P H E R A L I N P U T PATHWAYS P R O J E C T I N G TO THE MOTOR CORTEX IN T H E CAT
I-I. ASANUMA, K. D. LARSEN and P. ZARZECKI* The Rockefeller University, New York, N. Y. 10021 (U.S.A.)
(Accepted December 24th, 1978)
SUMMARY The possibility that the motor cortex receives peripheral input directly from the thalamus was examined using the evoked potential method and the following results were obtained. Potentials in the motor cortex evoked by stimulation of superficial radial (SR) or group II deep radial (DR) nerve were neither abolished nor delayed by ablation of the sensory cortex. Potentials in the motor cortex evoked by stimulation of group II D R nerve were most severely reduced by interruption of the spinocervical tract. Potentials evoked by stimulation of SR nerve were more severely reduced in the sensory cortex than in the motor cortex by section of the dorsal funiculus or cooling of the cuneate nucleus. The size of evoked potentials in the motor cortex increased rapidly when stimulus intensity to D R nerve exceeded the threshold to group II fibers. The results suggest that some inputs from the SR and group II D R nerves reach the motor cortex without a relay through the sensory cortex. INTRODUCTION Cells of the motor cortex receive topographically organized peripheral inputs 7, 11, but the pathway through which these inputs travel has not been fully defined. The motor cortex receives inputs from several cortical areas and subcortical nuclei. Cells of the thalamic nucleus ventralis lateralis which constitute the major input to motor cortex are not responsible for the somesthetic inputs because they do not receive precise information from the periphery3,21, 22. The sensory cortex projects to motor * Present address: Department of Physiology, Queen's University, Kingston, Ontario, Canada K7L 3N6.
198 cortex 1~ and is a possible source of the somesthetic inputs. This possibility was further emphasized by Wiesendanger 31, on the basis of the difference of lateneies for unit discharges in the motor and the sensory cortices. However, other evidence suggested that the sensory cortex cannot be the only source: peripherally evoked potentials in the monkey motor cortex do not disappear after ablation 2° or cooling 27 of the sensory cortex, and the latencies of the evoked potentials are not longer in the motor than in the sensory cortex 27. If there is a direct input from the periphery to the motor cortex, it may constitute the simplest cortical reflex pathway which could possibly be the basis of various cortical reflexes. Because of the potential importance of such a reflex, we wished to determine if the sensory inputs to the motor cortex arrive through the sensory cortex or if they arrive directly from the thalamus independent of the sensory cortex. This has been studied in several ways and is presented in this and the following two papers 4,19 Cats were used for the experiments since the relationships between afferent input and efferent outflow of the feline motor cortex 6 are similar to those in monkeys 27 and man 12, and because they are readily available. In the first series of experiments, the question of whether the evoked potentials in the motor cortex are dependent on the peripheral inputs which come through sensory cortex is examined by removing the sensory cortex or making selected spinal tract sections. It will be shown that some somesthetic inputs to the motor cortex do not come through the sensory cortex. These direct inputs to the motor cortex travel largely in the spinocervical tract whereas the inputs to the sensory cortex travel both in the spinocervical tract and the dorsal fasciculus. The thalamic nuclei responsible for the relay of the sensory input to motor cortex will be considered in the following two papers4, ~9. METHODS Experiments were performed on 14 cats of either sex weighing between 2.5 and 3.5 kg, anesthetized with pentobarbital sodium (Nembutal, 35 mg/kg). The pericruciate and coronal sulci on both sides were exposed by opening the skull and were covered by a mixture of mineral oil and vaseline. The superficial and deep radial nerves of both sides were dissected and mounted on bipolar stimulating electrodes in mineral oil pools made of skin flaps. The cervical cord was exposed from C-1 to C-4 by laminectomy and the dura was opened to provide access to dorsal or lateral funiculus or both during the experiments. In addition, in 6 cats the foramen magnum was enlarged and the caudal cerebellar vermis was partly removed to allow placement of a cooling probe on the surface of the caudal medulla over the cuneate nucleus.
Stimulating and recording Bipolar Ag-AgC1 electrodes were used to stimulate the deep radial (DR) and the superficial radial (SR) nerves. Single pulse stimuli were delivered at 2-3 sec intervals and the evoked potentials were recorded from the surface of the cerebral cortex by silver ball electrodes placed around the cruciate sulcus (motor cortex), coronal sulcus
199 (sensory cortex) and post-cruciate dimple (area 3a of Hassler and Muhs-Clement14). Potentials were recorded simultaneously from two cortical sites, and were averaged on-line (Fabri-Tek, model 1062) and displayed on an oscilloscope.
Cooling the euneate nucleus In 6 experiments a cooling probe with a ventral surface of 3 x 10 m m was positioned so that it was lightly touching the surface of the caudal medulla over the nuclei gracilis and cuneatus. While recording from the cortex, saline cooled to 0 °C was pumped through the probe.
Histological examination Whenever cortex was removed or the dorsal and/or lateral funiculi were cut, the cortex or cervical cord was removed at the end of the experiment and fixed in 10~o formalin. Frozen sections were made and stained by Kliiver and Barrera's method 16 to reconstruct the cut or lesioned area. RESULTS
Characteristics of cortical-evoked potentials The distribution of evoked potentials elicited by stimulation of superficial radial (SR) and deep radial (DR) nerves was examined first. Although this has been studied before, as summarized by Woolsey 82, the recent development of an on-line averaging computer made it possible to determine the distribution with greater accuracy. Two cats were used for this purpose and provided similar results. The evoked potentials were elicited by stimulation with two times the threshold strength which was supramaximal for cortical-evoked potentials. The potentials were recorded from 120 spots around the cruciate and coronal sulci, forming a grid of 1.0 m m squares. At each point, 32 evoked potentials were averaged and the voltage differences between the peaks of the initial positive phase and the following negative phase were measured as shown by the inset diagram in Fig. 1. Isopotential maps were constructed for both the
DR f F
Fig. 1. Distribution of evoked potentials elicited by stimulation of superficial radial (SR) and deep radial (DR) nerves. Supramaximal stimuli (3 x threshold stimuli) were used. The voltage difference between the positive and the negative waves were measured as shown by the inset diagram and equipotential points were connected at 50 #V steps, the outermost lines being 50/~V. Further details in the text.
200
Control
S Cx removed
I
__J ,oo,, 5msec
5reset
2rr
AS
,CrS
Fig. 2. SR- and DR-evoked potentials in the motor cortex before and after removal of the sensory cortex. Arrows on the evoked potentials indicate the arrival of the afferent volleys. The brain photograph and the sketch of histological slide show the extent of the aspiration. AS, ansate sulcus; CrS, cruciate sulcus; CoS, coronal sulcus. SR- and DR- (Fig. 1A, B) evoked potentials. In agreement with a study carried out under similar conditions 2z, SR stimulation elicited large evoked potentials at the lateral tip of the cruciate sulcus, lateral to the coronal sulcus, and at the post-cruciate dimple. D R stimulation elicited large potentials at the post-cruciate dimple and lateral to the cruciate sulcus but only small potentials lateral to the coronal sulcus as reported previously25, z0. In the next trial, the sensory cortex was removed as widely as possible in two cats to determine its possible contribution to the evoked potentials in the motor cortex. The results of both experiments were similar and one of them is illustrated in Fig. 2. The area of sensory cortex which was removed (inset photograph, Fig. 2) included the post-cruciate dimple (area 3a), coronal gyrus (area 3b, 1) and suprasylvian gyrus (area
201
A
B DR sttm
SR sl~m
o
100
o
"
100
~80
~
~oo
~ oo.
i
~.
,- 40 20 ~
8O
40.
•
13~T
o.--
-
"
_ _
!
/ "-,/I
20.
/...............~" 1.0
[
112
" 1.4 " 1.6 Stimulus Strength x T
---
~T
r-.,._ ~-
",
"
118
10
11
1.2 13 14 St~nulus Strength x
1.5 T
!~/ / ~ .
]%
"
I
v
:,,;.'"
Fig. 3. Relationships between stimulus intensity and the amplitude of evoked potentials for DR and SR nerves. A, response curves for DR stimulation. Specimen-evoked potentials are shown underneath the graph. Calibration pulses on the specimen records are 50 ttV, 5 msec. An arrow on the specimen (1.7 × T) indicates arrival of the incoming volley. B : curves for SR stimulation. Further details in the text. 2) although the depths of cortex around the coronal sulcus were spared. Recordings were made 3 h after removal of the sensory cortex. Both SR- and DR-evoked potentials became much smaller (about 1/3 of the control) after the removal. Furthermore, the shape of the SR-evoked potentials changed in both cases as shown in the upper traces. However, the latencies of both SR- and DR-evoked potentials were not altered by the lesion. It is known that the amplitude of the DR-evoked potential in area 3a is proportional to the amplitude of group I volleys in the D R nerve, and that an increase in D R stimulation beyond 1.6-2.0 times the threshold intensity for group I fibers does not increase the size of the DR-evoked potentials 2~. We examined the relationship between the intensity of the stimulus and the amplitude of DR-evoked potentials at the dimple (area 3a) and the m o t o r cortex. The experiments were carried out in 3 cats which provided similar results. The threshold strength for eliciting cortical-evoked potentials was taken as threshold strength for SR or D R nerve fibers as has been described previously 25,2s. The amplitude of DR-evoked potentials in area 3a increased rapidly as the stimulus intensity was increased from 1.0 to 1.4 times the threshold (Fig. 3A). In the m o t o r cortex, however, the amplitude started increasing rapidly at around 1.5 times the threshold. Since the recruitment of group II fibers starts at around 1.5 times the threshold strength for group I fibers is, the components of DR-evoked potentials recruited above 1.5 times the threshold are likely to be elicited by inputs
16
202
¢N
.;.:::,.
}
/
.. Ci o// / rn
~
.
i-:
.
gSUOdSe8
IOJ1UO~) %
¢0
,',|
1~
oY
!L
