- Email: [email protected]
Characteristics of the ipsilateral movement-related neuron in the motor cortex of the monkey
Brain Research, 204 (1981) 29-42 © Elsevier/North-Holland Biomedical Press
29
CHARACTERISTICS OF THE IPSILATERAL MO'VEMENT-RELATED N E U R O N I N T H E M O T O R C O R T E X OF T H E M O N K E Y
KEN'ICH MATSUNAMI and IKUMA HAMADA Department of Neurophysiology, Primate Research Institute, Kyoto University, lnuyama City, 484 (Japan)
(Accepted June 26th, 1980) Key words: motor cortex - - primate brain - - movement
SUMMARY The characteristics of the precentral neuron activity related to ipsilateral movements were studied while the monkey was performing finger, wrist and arm movements on either side. Out of 197 task-related neurons, 134 discharged in association with contralateral movements, but not with any one of 3 ipsilateral movements. Fifty neurons discharged with bilateral movements. Thirteen neurons discharged in assoeintion with ipsilateral movements 0psineurons). Ten were recorded from the trunk or shoulder area of the m o t o r cortex and were accompanied by contraction of those muscles by intracortical stimulation (ICS). The remaining 3 were related to elbow or wrist, but no ipsi-neurons were related to finger muscle contractions. In ipsilateral task performance, 7 ipsi-neurons discharged in association with finger and/or wrist movements in addition to a r m movement. Five others were associated with arm movement. The last one discharged with wrist movement. Most of the units showed similar response to contralateral movement. Ipsi-neurons were classified into two groups. One group was recorded around the sulcus precentralis superior, had the lower threshold current and was mostly associated with finger, wrist and arm movements. The other was recorded in the rostral m o t o r cortex, and had the higher threshold current and was related to arm movement. Among 185 neurons to which pyramidal tract stimulation was delivered, 2 out of the 80 PTNs and 11 out of the 105 non-PTNs were ipsi-neurons. E M G s were recorded from various muscles involved in the forelimb movements. A r m and finger muscles showed no activity when the monkey used the ipsilateral hand, while most of the shoulder and trunk muscles showed tonic or moderate
30 transient changes in the activity during the ipsilateral tasks. The ipsi-neuron activity was discussed in consideration with EMGs.
INTRODUCTION Ipsilaterally descending corticospinal or cortico-bulbo-spinal fibers originate mainly from the trunk and the proximal limb area of the motor cortex3,4,10,11,13,17. These ipsilateral fibers were considered to mediate ipsilateral control over the trunk and proximal extremity muscles 10, and to be important in the recovery of motor deficits after contralateral cortical damages s,ls. Physiologically, neurons related to ipsilateral movements exist in the motor cortex of man and monkeyT,L The anatomical findings suggested that neuron activity related to ipsilateral movement would be recorded from the trunk and proximal limb area of the motor cortex and would be associated with movements of these muscles, but this remains to be demonstrated. In addition, the number of neurons related to ipsilateral movements in the previous studies is very small 7,9. Consequently, the description of their characteristics is not satisfactory. Three points are examined in this report. First, to determine if the neurons related to ipsilateral movements are more associated with the trunk and proximal movement. Second, to establish if they are more recorded from the trunk and proximal limb area of the motor cortex. The third, to give more information about their characteristics. Previous short reports have been published 14-16. METHODS
Tasks and training procedures The experimental arrangement has been described in detail in a previous report 15, but it will be preferable to summarize several points for better understanding. Two rhesus monkeys were trained to perform 3 different tasks - - finger, wrist and arm movements - - with one hand following visual instructions presented on a control panel. The monkey was restrained in a monkey chair and faced the control panel, on which switches and lamps of light-emitting diodes (LEDs) were symmetrically arranged about the midline. For finger movement, the monkey depressed a small lever switch on the left side with the left index finger, when a LED above the switch was turned on (GO-signal). When a LED above a right switch was turned on, the monkey had to use the right hand. Every time the monkey closed a switch, another LED above GO-signal was illuminated, serving as a success signal. The wrist movement was conducted in a similar manner to finger movement, but the lever switch used was stiffer. In order for the monkey to perform arm movement, a handle was provided. The position of the handle was at the same horizontal level as the elbow joint of the monkey. As instruction cues for arm movement, two sets o f 3 LEDs in a row were
31 inlaid at the upper part of the control panel, one on the left and the other on the right. The monkey moved the handle in arc according to the visual cues. The side of the hand used in a task matched the side of the lamps illuminated on the panel. The movement direction of the handle was indicated by two of 3 LEDs in a set. When the outer L E D was turned on, the monkey had to move the handle outwards (outwards here means the direction from the midline toward a lateral side), and held it in the outer target zone (15 °) for half a second. Then the inner target LED was turned on, telling the monkey to move it back into the inner target zone. The distance between the outer and inner target zones was 40 ° in arc. Success was signaled by lighting the third LED between the two target LEDs, with a sound cue. Displacement of finger and wrist movements were detected with strain gauges glued on the lever of each switch. The position of the arm was converted into a electric signal through a potentiometer connected to the pivot of the handle. In each task, the monkey was rewarded with a drop of juice after 9 sequential successes (fixed ratio).
Surgery After completion of the training for 4 months, the monkey underwent surgery with pentobarbital anesthesia under aseptic conditions. Removing a piece of the skull appropriately over the forelimb area of the motor cortex, a cylinder (18 mm i.d.) was fixed to the bone, centered at the coordinates (A = 14, L = 15). It served as a mount for a hydraulic manipulator (Narishige, MO-8). A bipolar stimulating electrode of tungsten wires (300/~m in diameter) coated with polyurethan resin, was implanted in the pyramidal tract (A = 0) for identification of pyramidal tract neurons (PTNs).
Recording session After the monkey's recovery from surgery, daily recordings were started. Neuron activity was recorded with a glass-coated Pt-Ir microelectrode (1-4 Mf~ for 60 Hz). Electrode penetrations were made in the motor cortex on a 1-mm grid. The activities were audiomonitored and only that which related to a movement was recorded. At the end of each microelectrode penetration, the electrode was drawn back upwards once more to the site where units had been recorded and intracortical stimulation (ICS) was delivered (10 pulses of 0.1 msec duration at 100 Hz, every 1 sec). Muscle contractions were observed visually and tactilely. The amount of current varied from 7/~A for finger movement to over 500 #A for trunk and proximal muscles. The larger amount of current used for trunk and proximal limb muscles compared to the previous study 12 could be explained by several factors. First, deep truncal and shoulder muscles could be detected from the surface after suprathreshold contraction. Second, the points to which the electrodes were drawn back after unit recording were not the best points for stimulation. Third, the present aim of the ICS was to determine exactly what muscles were moved, and for this purpose, sometimes current a little stronger than the threshold, was delivered. But in order to avoid confusion with the foregoing studies1, TM, we used the term intracortical stimulation (ICS) instead of intracortical microstimulation, though we followed the same methods 1. Recordings of neuron activity, displacement of finger, wrist and arm movements and voices were stored on a 4-channel F M magnetic tape.
32
Data analysis Stored data were played-back for off-line analysis with a PDP-12 computer. Average time histograms were constructed for all 6 tasks. The analysis epochs were comprised of 50 bins of 10 msec for finger and wrist movements, and 20 msec for arm movements. Neuron activity was sampled for 25 bins before and after the start of each movement and was averaged for 48 trials. Though the neuronal activity recorded was related to, at least, one of the 6 tasks, we used statistics to grade their activities with movements. The 50 bins were put into 6 sets in sequence from the first bin. The first and the last sets were composed of 9 bins and the remaining 4 were composed of 8 bins. Mean discharge rate in each of the 6 sets was calculated. Then the F-test was applied to detect whether all 6 mean values were equal or not. If the null hypothesis was rejected, the unit was considered to change its discharge rate in association with movement. Change of activity was graded as I ( + ) , II ( + + ) and III ( + + ÷ ) according to the significance levels of 1 ~o, 0.5 ~o and 0.1 respectively. A 1 ~o significance was adopted to contralateral movement, and 0.I ~o for ipsilateral movement. A severe criterion was held to ipsilateral movement in order to pick up a neuron that was surely related to it. The time of activity change corresponding to a movement was determined by eye on a histogram, where 3 successive bins increased (or decreased) sequentially.
EMG recordings Sometimes during the daily recording sessions, EMGs of muscles involved in forelimb movements were recorded in the identical situation to the unit study. A bipolar electrode, made of thin copper wires coated with enamel except for the tips, was inserted in various muscles accessible and identifiable from the surface. The recordings were repeated at least twice for each muscle on different occasions. In the case of a large muscle such as m. latissimus dorsi, EMGs were obtained from several different points. The upper part of this muscle was involved in all forelimb movements, while the lower part was almost inactive. Muscular activity was processed int o averaged histogram in a similar way to neuron activity. The amplitude of muscular activity was proportionally converted into the height in each bin of a histogram, using a comparator to discriminate a signal from noise. Forty-eight trials were averaged. Bin width was 10 msec for finger and wrist movements and 20 msec for arm movement. The following abbreviations were used for muscles; BC, m. biceps; BR, m. brachioradialis; DEX, deep extensor of the forearm; DFX, deep flexor of the forearm; DLT, m. deltoideus; ECR, m. ext. carpi radialis; FCR, m. flex. carpi radialis; ISP, m. infraspinatus; LAT, m. lattisimus dorsi; PCT, m. pectoralis major; RHM, m. rhomboideus; SCA, m. splenius capitis; SER, m. serratus ant.; SSP, m. supraspinatus; TC, m. triceps; TMA, m. terres major; TMI, m. terres minor; TRP, m. trapezius; 34F, genuine finger muscle between 3rd and 4th digits.
33 RESULTS
Recordings have been made from 197 neurons in the motor cortex of two monkeys. Sampling bias was carefully avoided for two points. First, neuronal activities were evenly recorded from the trunk and proximal limb area as well as from the distal limb area in the motor cortex. Second, any neuronal activity related to ipsilateral or contralateral movement was recorded. Out of 197 units, the discharge of 134 were associated with contralateral tasks, but not with any one of 3 ipsilateral tasks (contra-neurons). Sixty-three units discharged in relation to, at least, one of 3 ipsilateral tasks in addition to contralateral tasks. To these 63 neurons, the following criteria were applied to classify them into two groups. Criteria. (1) The peak and mean frequency during 50 bins around the onset o f movement were greater for ipsilateral than for contralateral movement and were greater than 10 spikes/sec.
A F
IPSI
CONTRA
tltllllll,i,hhl,hh,l,,I,l,,ll,,llhlll~,l~
w
,
,
(;
..
'
F
,
,,,,I,Itl.l,,,,....,..',,,,.l,.,,h.,.,h,,h~]
w
,
,
~
~
.
,',.~.___.
'" .ht,,,,,,,,h,ldldldDh~IHhhlllhlI
'"
i
I
llllilili,,,,,i,li,,ihl,,,l,ii,llillll OUTl ,,!,,,,.,,.,,~...,,,.i ', ,,dl,i,lhIlSli~l l
OUT
'
~
,
,
I B F
I PSI
hlLhl.
,. I L .
CONTRA
,
hi I l i l l l h l l l . h l l h l
,ill.
]
'.
I
.*'
, .
;"
]
..;
- ~ - - - I
W
i 0 I
IN ............,11.........!lllllll,!,,.,I....u ] .
.
.
.
.
IN
i
OUTi,,h,h,ll.,,.,,.,mL
,~ , , , I . h , I ,,., ,*,1 u
.
.
i, I
OUT
, o
Fig. 1. Example o f two ipsi-neurons in averaged time histograms. A a n d B: different neurons. The left column, labelled as IPSI, shows the histograms for ipsilateral finger (F), wrist (W), a r m m o v e m e n t in inward (IN) and outward (OUT) directions. The right column, labelled as C O N T R A , shows contralateral movements. A vertical bar represents 40 ° for arm displacement. A horizontal bar represents 200 msec for a r m and 100 msec for finger and wrist movements.
34
ant.
~
L f H s@E E(~S sH f f WSWW~S S B H f ~WWFwSWSF FWWWw(~B)WBB/~ .(~pfSBr .___s B e
--
~WW
E F(E)C S
~ . ~ w F W~)B
~B N N ~ a
Y
s.~"s N f
WwsSSSS
~eFFE FF S/'" F FWSF. I f W S W(t~,S W S~
E
~e°t
post.
Fig. 2. Representation of ipsi-neurons on the motor cortex. Figure shows two motor areas surrounded by the sulcus arcuatus (s.arc) anteriorly, by the sulcus centralis (s.cent) posteriorly, and by the sulcus precentralis superior (s.pre) medially. Each capital letter indicates the entrance of an electrode and the abbreviated name of a part of the body or forelimb. Encircledletters showthe locations ofipsi-neurons. Abbreviation; B, back; C, chest; E, elbow; F, finger; f, foot; H, hip; L, loin; 1,lip; N, neck; S, shoulder; W, wrist. (2) The unit activity started to increase 100 msec or earlier before the onset of ipsilateral movement. (3) The period from the onset of increase in unit discharge to the onset of the movement was longer for ipsilateral than for contralateral movement. Thirteen neurons completed all 3 criteria and they were defined as 'ipsi-neurons'. The other 50 did not complete all 3. They were defined as 'bilateral neurons'. In this report, description was restricted to the characteristics of the 13 ipsi-neurons. Fig. 1 shows examples of two ipsi-neurons in averaged time histograms. One neuron (Fig. 1A) changed its activity in all task performance on both sides. In the finger tasks (F), the activity decreased with the downward swing of the movements during ipsilateral and contralateral performances, but the change was more distinct in the ipsilateral movement. In the wrist tasks (W), the change was more remarkable in contralateral than in ipsilateral movement. For arm movements, the activity increased during ipsilateral inward movement (IN) and decreased during outward movement (OUT). This unit satisfied the 3 criteria for the ipsilateral inward movement. For the contralateral arm movement, the activity transiently increased then decreased during inward movement and gradually increased during outward movement. This neuron was fast P T N (0.7 msec antidromic latency) and ICS (47/~A) produced the contraction of shoulder muscles. In Fig. 1B, another ipsi-neuron is illustrated. It fired in all 3 ipsilateral movements and it was most distinct in the ipsilateral inward arm movement (IN). But, the unit was almost silent in all 3 contralateral tasks. This neuron was non-PTN and ICS (143 /,A) evoked the contraction of contralateral shoulder muscles. The response pattern of the remaining 11 ipsi-neurons is summarized in Table I with their other properties,
35
Location of ipsi-neurons in the motor cortex From histological data4,10,11, it could be inferred that ipsi-neuron activity could be more recorded in the trunk and proximal limb area in the motor cortex. To ascertain this inference, locations of ipsi-neurons were mapped on the motor cortex. In Fig. 2, capital letters denote the entrances of microelectrodes in a 1-mm grid on the motor cortex. The capital letters also represent the abbreviated names of parts of the body or the forelimb, identified by muscle contraction with ICS. The full name represented by each capital letter is given in the legend. The 12 encircled capital letters signify the tracks at which ipsi-neurons were encountered. Described in detail, 7 ipsineurons were obtained from 7 shoulder (S) tracks where ICS contracted shoulder muscles. Similarly, 1 ipsi-neuron was obtained from 1 back (B) track, 2 from one chest (C), 1 from 1 elbow (E), and 2 from 2 wrist (W) tracks. No ipsi-neurons were obtained from finger (F) tracks. Fig. 2 also shows that ICS contracted muscles in the lip (1), neck (N), hip (H), loin (L) and foot (f) in 17 tracks. But most neurons in these areas were not associated with any task performances. Six neurons were, however, related to one of the 6 tasks, one of them being bilateral neuron. The other 5, recorded in the hip, loin, or foot area, were contralateral units. Correlation of ipsi-neurons with muscular contractions To ascertain the foregoing inference further, the appearance rate of ipsineurons in the finger (F), wrist (W), elbow (E), shoulder (S) and trunk (T) tracks were separately studied. The chest and back were here included in the trunk. Six neurons related to muscular contraction of the lip, neck, loin, hip or foot were excluded. For the remaining 191 neurons, the following figures were obtained. Three ipsi-neurons (12.0~) were found in 25 T-neurons. Similarly, 7 ipsi-neurons (9.3 ~o) in 75 S-units, one (3.3 %) in 30 E-units, and 2 (6.3 %) in 32 W-units were also found. No ipsi-neurons were found in 29 F-units. Therefore, it was concluded that ipsi-neurons appeared with a higher rate in units related to trunk and proximal limb muscles. If T- and S-neurons were united into a 'proximal' group, and F-, W-, and E-neurons were categorized as a 'distal' group, the percentage of ipsi-neurons was 10 % (10/100) in the proximal and 3.3 % (3/91) in the distal group with a significant difference by z~-test (P < 0.05). Response pattern of ipsi-neurons to 6 movements Concerning the firing pattern of ipsi-neurons, it would be interesting to know the correlation of activities between finger, wrist or arm movement ipsilaterally, or the correlation between the ipsilateral and the contralateral activity. Table I summarizes this relation in 6 tasks. Five ipsi-neurons from the rostral motor cortex (RST-group) are shown in the upper part and 7 recorded near the sulcus precentralis superior (SPSgroup) are shown in the lower part of the Table. Four out of 5 RST-units responded to ipsilateral and contralateral arm movements, but neither to finger nor wrist movement. ICS contracted shoulder muscles in 3 of these 4 ipsi-units and back muscles in the remaining one ipsi-unit. One RST-unit was associated with ipsilateral and contralateral arm and finger movements, but the activity was more associated with ipsilateral movements. Thus RST ipsi-neurons were related to the arm rather than distal movements.
36
÷+÷÷ ÷÷~÷÷ ~÷÷÷÷ Z N ~
÷ ÷
~
÷÷~-~÷ ÷÷÷÷-~
It+++++ ÷ ÷
÷÷ ÷÷,÷÷
+l
I I I l l l
++++
÷÷÷
0
III1+
+
÷÷÷÷ ÷÷÷÷÷ • ÷÷÷÷ ÷÷÷÷÷
÷ ÷ +
÷
I I I I I +
÷ 0
I I I I+
÷
+
II
I+++
±÷÷~÷ -~-÷÷÷÷÷ ÷-~÷÷÷÷
I++++++
÷÷÷
÷÷÷
+++1+++ ÷÷ ÷÷
I++1+++
0
r~
r!. i
~
.~
o
÷ -~
~.
÷ ~-
37 O n the other hand, 5 o f the 7 SPS-units were associated with ipsilateral finger and wrist movements in addition to a r m movement. Three o f these 5 were also associated with contralateral finger, wrist and a r m movements. The fourth was not associated with any contralateral movement. The last one o f the 5 was related only to a r m m o v e m e n t contralaterally. But the activity was less significant than for the ipsilateral movement. O f the remaining two SPS-units, one was associated with ipsilateral a r m and with contralateral wrist and a r m movements. The other one was associated with only wrist movements on both sides. ICS produced wrist m o v e m e n t o f this neuron. The a m o u n t o f threshold current for ICS was different between SPS-group (252 -t- 132/zA) and R S T - g r o u p (542 4- 181 p A ) with statistical significance (P < 0.01). This reflected the fact that the a m o u n t o f the threshold current was smaller near the sulcus precentralis superior than in the rostral part o f the m o t o r cortex 20. Effects o f ICS were primarily contralateral to the stimulation. Ipsilateral muscle contractions were observed, but at that time contralateral contractions were as strong or dominant, and isolated ipsilateral muscular contraction was never observedS, e.
Ipsi-neurons and PTNs There are direct and indirect ipsilateral pathways to the spinal cord z0. F r o m this point, it would be interesting to k n o w the distribution o f ipsi-neurons in P T N and IPSI
CONTRA IA
IF
IW
IN
CA OUT
CF
34 F
CW '
~
IN ~ -
OUT
~lk
DEX OFX 8C
TC
BR
6
6
0
b
0
0
6
0
Fig. 3. EMGs of distal muscles. CONTRA: EMGs recorded from contralateral muscles during contralateral task performances. IPSI: EMGs recorded from contralateral muscles during ipsilatcral tasks. CF, CW, CA denote contralateral finger, wrist and arm movements. IF, IW, IA denote ipsilateral finger, wrist and arm movements. IN and OUT indicate inward and outward arm movements. BC, m. biceps; BR, m. brachioradialis; DEX, deep extensor muscle of the forearm; DFX, deep flexor of the forearm; ECR, m. ext. carpi radialis; FCR, m. flex. carpi radialis; TC, m. triceps; 34F, genuine finger muscles between 3rd and 4th fingers. Time scale, 250 msec for finger and wrist movements and 500 msec for arm movement before and after the onset of movements. A vertical bar on the right represents 100 pV.
38
IPSI IF
IW
CONTRA
IN
IA OUT
CF
CW
IN
CA
OUT
TRP
SER
I I
mmmm m
il
.,,, . - -
TM f
RHM
$sP
L SCA 0
0
0
0
0
O
o
o
Fig. 4. EMGs of proximal muscles. Explanation is the same as in Fig. 3. Abbreviation: DLT, m. deltoideus; IPS, m. infraspinatus; LAT, m. latissimus dorsi; PCT, m. pectoralis major; RHM, m. rhomboideus; SCA, m. splenius capitis; SER, m. serratus anterior; SSP, m. supraspinatus; TMA, m. terres major; TMI, m. terres minor; TRP, m. trapezius; Time scale, 250 msee for finger and wrist movements and 500 msee for arm movement before and after the onset of movements. Vertical bars represent 100 #V. non-PTN. The results were as follows; two (2.5 %) out of 80 PTNs and 11 (10.5 ~ ) out of 105 non-PTNs were ipsi-neurons. For the rest of neurons, the antidromic stimulation was not tested. These two ipsi-PTNs were fast-PTNs (Table I). The border latency to separate the fast and the slow was 2.0 msec zS. Moreover, ipsi-neurons were equally found in phasically as well as in tonically firing neurons, and in neurons with low spontaneous discharge rate as well as with higher frequency.
EMG during ipsilateral and contralateral tasks In consideration of ipsi-neurons, it is important to refer to muscular activities during ipsilateral and contralateral tasks. Fig. 3 a n d 4 are presented for this purpose. A time histogram of E M G was made in a similar way to unit study (cf. Methods). E M G s were recorded from the muscles contralateral to the recording cortex during the contralateral ( C O N T R A ) and ipsitateral (IPSI) task performances. Fig, 3 shows E M G s of 'distal' muscles. F o r all the muscles in this group, no significant activity was observed, when the monkey used the ipsilateral forelimb. Activity of biceps and triceps muscles was very weak during the task on either side. All other 'distal' muscles
39 changed their activities in, at least, one of 3 contralateral tasks, especially with the wrist. But it is necessary to comment on the EMG of finger muscles. When a EMG was recorded from the muscles between the 3rd and 4th fingers, as shown in Fig. 3, the activity was negligible during both ipsilateral and contralateral finger tasks. This reflected the performance of the monkey. During the contralateral finger task, the monkey depressed the small lever with index finger flexion, the 3rd and 4th fingers being kept in a loosely flexed state and out of use. When a EMG was recorded from the intrinsic finger muscles between the 2nd and 3rd digits, moderate activity was recorded during ipsilateral finger movement, though no activity was observed during the ipsilateral wrist movement. At that time, the monkey rested the contralateral hand quietly on a rod before its chest. Therefore, we could not suppress the activity of the contralateral finger muscles between the 2nd and 3rd digits during the ipsilateral finger task. In Fig. 4, the activity of trunk and 'proximal' muscles is illustrated. Most of the muscles in this group changed activity with contralateral finger (CF) and wrist (CW) movement. Changes of activities were also observed during the contralateral arm movements in the following muscles; mm. deltoideus (DLT), trapezius (TRP), serratus ant. (SER), terres major (TMA), pectoralis major (PCT) and infraspinatus (ISP). During ipsilateral task performances, activity change was detected in TRP, TMA, rhomboideus (RHM) and ISP muscles. Some muscles in this group were tonically active during the task with ipsilateral arm (IA). Activities of TRP, PCT and supraspinatus (SSP) had characteristics of 'distal' muscles in that they discharged phasically during CF and CW movements and were inactive during ipsilateral finger (IF) and wrist (IW) movements. Consideration of EMGs in relation to ipsi-neuron activity will be taken into account in the Discussion. DISCUSSION
Validity of criteria of ipsi-neurons The 3 criteria were adopted to define an ipsi-neuron (of. Results). These criteria seem to be severe, but still a question was raised. Are ipsi-neurons defined as such actually associated with the ipsilateral muscular activities, or is the change of activity in ipsilateral movement a secondary effect due to the contamination from contralateral movement? The discussion will be focused on this point in consideration with EMGs. For the 'distal' muscles, the contamination from contralateral movements can be neglected, because 'distal' limb muscles did not show any significant activities during ipsilateral performances. But in the case of the trunk and 'proximal' muscles, the result must be carefully interpreted. Some of the contralateral muscles in this group changed their activities in the ipsilateral tasks. As shown in Fig. 4, activities of the deltoideus (DLT), trapezius (TRP), terres major (TMA), rhomboideus (RHM), infraspinatus (ISP), and splenius capitis (SCA) muscles were modulated during ipsilateral arm movement. Activities during ipsilateral
40 tasks may possibly be due to the contamination from the contralateral movements. But this possibility can be excluded for the TRP and ISP muscles, because these muscles showed very significant change of activity during contralateral movements. If an ipsi-neuron was associated with these muscles, its activity was to be more related to contralateral than to ipsilateral movement. The SCA muscle can be excluded because it is a neck muscle and no ipsi-neuron activity was recorded from the neck area in the motor cortex. The DLT muscle was excluded because its activity was hardly modulated during ipsilateral task performances. But the TMA and R H M muscles could not safely be neglected only from the comparison of temporal pattern of EMGs and neuron activity. But the activities of these muscles during contralateral finger and wrist tasks were much more distinct than during ipsilateral tasks. Ipsi-neurons did not show such change of activity when they were associated with ipsilateral finger or wrist movement. Taken all together, it is more likely that ipsi-neurons were associated with ipsilateral rather than with contralateral movements.
Functional consideration of ipsi-neurons Based on the experimental and clinical data2-4,s,10,11, is, it was supposed that neuron activity associated with ipsilateral movement would be more related to trunk and proximal limb muscles. This supposition was confirmed by the present unit study. First, 12 out of the 13 ipsi-neurons were associated with ipsilateral arm movement and only one was solely related to wrist movement. Second, ICS provoked contraction of the trunk or shoulder muscles for 10 units and of elbow or wrist for the other 3 in their recorded locations. Third, the appearance rate of ipsi-neurons was higher in the proximal than in the distal group. In the present study, ipsi-units were mainly recorded in two different parts of the motor cortex; in the rostral (RST) motor cortex and near the sulcus precentralis superior (SPS). Units of two groups seem to have a little different function over ipsilateral movement. The RST-group was associated with ipsilateral arm movement, but the SPS-group was related to finger and/or wrist, in addition to arm movement. A somatotopical map 19 demonstrated that the RST-area corresponded to the trunk, and the SPS-area corresponded to the proximal limb muscles. Therefore RST-units would be responsible for ipsilateral regulation of the trunk or the control of the posture4,1°, while SPS-unit would be involved in ipsilateral proximal limb movement or limb-body integration4,1o. But the total number of ipsi-neurons was too small to attribute them the whole control of ipsilateral movement. Therefore, it seems more likely that ipsi- and bilateral neurons work in a cooperative manner to perform ipsilateral motor control, because the sum of ipsi- and bilateral neurons was 63 out of 197 movement-related neurons, constituting a fairly large number of the total. Comparison with the previous studies Three (7.5 %) out of 40 PTNs were associated with ipsilateral wrist extension 7. This figure was quite similar to the present result of 6.5 %. But if PTN and non-PTN were separately considered, a discrepancy arose. In the present experiment, 2 (2.5 %)
41 out of the 80 PTNs and 11 (10.5 ~ ) out of the 105 non-PTNs were ipsi-neurons. The ratio of ipsi-PTNs was very small in the present study. The discrepancy could be explained by the different criteria adopted in each report. The criterion previously reported 7 was the same as our first criterion, hence our criteria would be more severe than in the previous study, yielding a smaller ratio in the present report. The difference of the tasks might be another source of the discrepancy. Wrist extension was employed in the previous experiment 7, while we adopted 3 tasks using finger, wrist and arm. This procedure would enlarge the area from which ipsi-neuron activity was favoured to be recorded, by including the rostral motor cortex. Therefore, the difference of criteria may be the major reason for the discrepancy. Ipsi-neurons were significantly more numerous in n o n - P T N than in PTN. It is doubtful that the majority of PTNs were sampled from the distal limb area of the motor cortex, from which only a smaller portion of ipsilateral corticofugal fibers originateda,4,10,11. But this was to be excluded because 37 out of the 80 PTNs were sampled from the distal motor area TM.Probably very small number of ipsi-PTNs in the parent population would cause an exaggerated result. But still the fact may be taken as a demonstration of the importance of indirect ipsilateral pathways over the regulation of ipsilateral movement. ACKNOWLEDGEMENT The authors thank Prof. K. Kubota for his encouragement in the course of the experiment and Mrs. T. Miwa for her technical assistance and Miss S. Ishikara for the typing. REFERENCES 1 Asanuma, H. and Sakata, H., Functional organization of a cortical efferent system examined with focal depth stimulation in cats, J. NeurophysioL, 30 (1967) 35-54. 2 Brinkman, J. and Kuypers, H. G. J. M., Split brain monkey: cerebral control of contralateral and ipsilaterai ann, hand and finger movements, Science, 176 (1972) 536-539. 3 Brinkman, J. and Kuypers, H. G. J. M., Cerebral control of contralateral and ipsilateral ann, hand and finger movements in the split-brain rhesus monkey, Brain, 96 (1973) 653-674. 4 Brinkman, J., Split Brain Monkey: Cerebral ControlofContralateralandlpsilateralArm, Handand Finger Movements, Doctoral Thesis, Univ. of Rotterdam, 1974. 5 Bucy, P., Representation of ipsilateral extremities in the cerebral cortex, Science, 78 (1933) 418. 6 Bucy, P. and Fulton, J. F., Ipsilateral representation in the motor cortex of monkeys, Brain, 56 (1933) 318-342. 7 Evarts, E. V., Pyramidal tract neuron activity associated with a conditioned hand movement in the monkey, J. Neurophysiol., 29 (1966) 1011-1027. 8 Glee, P., Contra and ipsilateral motor and sensory representation in the cerebral cortex of monkey and man. In K. J. Ziilich, O. Creutzfeldt and G. C. Galbraith (Eds.), Cerebral Localization, Berlin, Heidelberg, New York, 1975, pp. 48-61. 9 Goldring, S. and Racheson, R., Human motor cortex: sensory input data from single neuron recordings, Science, 175 (1972) 1493-1495. 10 Kuypers, H. G. J. M., The descending pathways to the spinal cord, their anatomy and function. In J. C. Eccles and J. C. Schade (Eds.), Progress in Brain Research, VoL I 1, Organization of The Spinal Cord, 1964, pp. 188-202. 11 Kuypers, H. G. J. M. and Brinkman, J., Precentral projections of different parts of the spinal intermediate zone in the rhesus monkey, Brain Research, 24 (1970) 29-48.
42 12 Kwan, H. C., MacKay, W. A., Murphy, J. T. and Wong, Y. C., Spatial organization of precentral cortex in awake primates. II. Motor outputs, J. Neurophysiol., 41 (1978) 1120-1131. 13 Liu, C. N. and Chambers, W. W., An experimental study of the cortico-spinal system in the monkey (Macaea mulatta). The spinal pathways and preterminal distribution of degenerating fibers following discrete lesions, J. comp. Neurol., 123 (1964) 257-284. 14 Matsunami, K. and Hamada, I., Motor cortex unit activity related to ipsilateral upper limb movement of awake monkeys, XXVII int. Congr. PhysioL Sci. (Paris), (1977) 489. 15 Matsunami, K. and Hamada, I., Precentral neuronal activities associated with the upper limb movement in monkeys, J. Physiol. (Paris), 74 (1978) 319-322, 16 Matsunami, K. and Hamada, I., Antidromiclatency of the monkey pyramidal tract neurons related to ipsilateral hand movements, Neurosci. Lett., 16 (1980) 245-249. 17 Tigges, J., Nakagawa, S. and Tigges, M., Efferents of area 4 in a South American monkey (Saimiri). L Terminations in the spinal cord, Brain Research, 171 (1979) 1-10. 18 Twitehell, T. E., The restoration of motor function following hemiplegia in man, Brain, 74 (1951) 443-480. 19 Woolsey, C. N., Organization of somatic sensory and motor area of the cerebral cortex. In H. F. I-Iarlowand C. N. Woolsey ('Eds.), Biological and BiochemicaI Basis o f Behavior, Univ. of Wisconsin Press, Madison, Wise., 1958, pp, 63-81. 20 Woolsey, C. N., Settlage, P. H., Meyer, D. R., Senccr, W., Pinto, T. and Travis, A. M., Patterns of localization in precentral and 'supplementary' motor areas and their relation to the concept of a premotor area. Chap. XH. In Patterns of Organization in The Central Nervous System, Res. Publ. Ass. nerv. ment. Dis., 30 (1951) 238-264.
29
CHARACTERISTICS OF THE IPSILATERAL MO'VEMENT-RELATED N E U R O N I N T H E M O T O R C O R T E X OF T H E M O N K E Y
KEN'ICH MATSUNAMI and IKUMA HAMADA Department of Neurophysiology, Primate Research Institute, Kyoto University, lnuyama City, 484 (Japan)
(Accepted June 26th, 1980) Key words: motor cortex - - primate brain - - movement
SUMMARY The characteristics of the precentral neuron activity related to ipsilateral movements were studied while the monkey was performing finger, wrist and arm movements on either side. Out of 197 task-related neurons, 134 discharged in association with contralateral movements, but not with any one of 3 ipsilateral movements. Fifty neurons discharged with bilateral movements. Thirteen neurons discharged in assoeintion with ipsilateral movements 0psineurons). Ten were recorded from the trunk or shoulder area of the m o t o r cortex and were accompanied by contraction of those muscles by intracortical stimulation (ICS). The remaining 3 were related to elbow or wrist, but no ipsi-neurons were related to finger muscle contractions. In ipsilateral task performance, 7 ipsi-neurons discharged in association with finger and/or wrist movements in addition to a r m movement. Five others were associated with arm movement. The last one discharged with wrist movement. Most of the units showed similar response to contralateral movement. Ipsi-neurons were classified into two groups. One group was recorded around the sulcus precentralis superior, had the lower threshold current and was mostly associated with finger, wrist and arm movements. The other was recorded in the rostral m o t o r cortex, and had the higher threshold current and was related to arm movement. Among 185 neurons to which pyramidal tract stimulation was delivered, 2 out of the 80 PTNs and 11 out of the 105 non-PTNs were ipsi-neurons. E M G s were recorded from various muscles involved in the forelimb movements. A r m and finger muscles showed no activity when the monkey used the ipsilateral hand, while most of the shoulder and trunk muscles showed tonic or moderate
30 transient changes in the activity during the ipsilateral tasks. The ipsi-neuron activity was discussed in consideration with EMGs.
INTRODUCTION Ipsilaterally descending corticospinal or cortico-bulbo-spinal fibers originate mainly from the trunk and the proximal limb area of the motor cortex3,4,10,11,13,17. These ipsilateral fibers were considered to mediate ipsilateral control over the trunk and proximal extremity muscles 10, and to be important in the recovery of motor deficits after contralateral cortical damages s,ls. Physiologically, neurons related to ipsilateral movements exist in the motor cortex of man and monkeyT,L The anatomical findings suggested that neuron activity related to ipsilateral movement would be recorded from the trunk and proximal limb area of the motor cortex and would be associated with movements of these muscles, but this remains to be demonstrated. In addition, the number of neurons related to ipsilateral movements in the previous studies is very small 7,9. Consequently, the description of their characteristics is not satisfactory. Three points are examined in this report. First, to determine if the neurons related to ipsilateral movements are more associated with the trunk and proximal movement. Second, to establish if they are more recorded from the trunk and proximal limb area of the motor cortex. The third, to give more information about their characteristics. Previous short reports have been published 14-16. METHODS
Tasks and training procedures The experimental arrangement has been described in detail in a previous report 15, but it will be preferable to summarize several points for better understanding. Two rhesus monkeys were trained to perform 3 different tasks - - finger, wrist and arm movements - - with one hand following visual instructions presented on a control panel. The monkey was restrained in a monkey chair and faced the control panel, on which switches and lamps of light-emitting diodes (LEDs) were symmetrically arranged about the midline. For finger movement, the monkey depressed a small lever switch on the left side with the left index finger, when a LED above the switch was turned on (GO-signal). When a LED above a right switch was turned on, the monkey had to use the right hand. Every time the monkey closed a switch, another LED above GO-signal was illuminated, serving as a success signal. The wrist movement was conducted in a similar manner to finger movement, but the lever switch used was stiffer. In order for the monkey to perform arm movement, a handle was provided. The position of the handle was at the same horizontal level as the elbow joint of the monkey. As instruction cues for arm movement, two sets o f 3 LEDs in a row were
31 inlaid at the upper part of the control panel, one on the left and the other on the right. The monkey moved the handle in arc according to the visual cues. The side of the hand used in a task matched the side of the lamps illuminated on the panel. The movement direction of the handle was indicated by two of 3 LEDs in a set. When the outer L E D was turned on, the monkey had to move the handle outwards (outwards here means the direction from the midline toward a lateral side), and held it in the outer target zone (15 °) for half a second. Then the inner target LED was turned on, telling the monkey to move it back into the inner target zone. The distance between the outer and inner target zones was 40 ° in arc. Success was signaled by lighting the third LED between the two target LEDs, with a sound cue. Displacement of finger and wrist movements were detected with strain gauges glued on the lever of each switch. The position of the arm was converted into a electric signal through a potentiometer connected to the pivot of the handle. In each task, the monkey was rewarded with a drop of juice after 9 sequential successes (fixed ratio).
Surgery After completion of the training for 4 months, the monkey underwent surgery with pentobarbital anesthesia under aseptic conditions. Removing a piece of the skull appropriately over the forelimb area of the motor cortex, a cylinder (18 mm i.d.) was fixed to the bone, centered at the coordinates (A = 14, L = 15). It served as a mount for a hydraulic manipulator (Narishige, MO-8). A bipolar stimulating electrode of tungsten wires (300/~m in diameter) coated with polyurethan resin, was implanted in the pyramidal tract (A = 0) for identification of pyramidal tract neurons (PTNs).
Recording session After the monkey's recovery from surgery, daily recordings were started. Neuron activity was recorded with a glass-coated Pt-Ir microelectrode (1-4 Mf~ for 60 Hz). Electrode penetrations were made in the motor cortex on a 1-mm grid. The activities were audiomonitored and only that which related to a movement was recorded. At the end of each microelectrode penetration, the electrode was drawn back upwards once more to the site where units had been recorded and intracortical stimulation (ICS) was delivered (10 pulses of 0.1 msec duration at 100 Hz, every 1 sec). Muscle contractions were observed visually and tactilely. The amount of current varied from 7/~A for finger movement to over 500 #A for trunk and proximal muscles. The larger amount of current used for trunk and proximal limb muscles compared to the previous study 12 could be explained by several factors. First, deep truncal and shoulder muscles could be detected from the surface after suprathreshold contraction. Second, the points to which the electrodes were drawn back after unit recording were not the best points for stimulation. Third, the present aim of the ICS was to determine exactly what muscles were moved, and for this purpose, sometimes current a little stronger than the threshold, was delivered. But in order to avoid confusion with the foregoing studies1, TM, we used the term intracortical stimulation (ICS) instead of intracortical microstimulation, though we followed the same methods 1. Recordings of neuron activity, displacement of finger, wrist and arm movements and voices were stored on a 4-channel F M magnetic tape.
32
Data analysis Stored data were played-back for off-line analysis with a PDP-12 computer. Average time histograms were constructed for all 6 tasks. The analysis epochs were comprised of 50 bins of 10 msec for finger and wrist movements, and 20 msec for arm movements. Neuron activity was sampled for 25 bins before and after the start of each movement and was averaged for 48 trials. Though the neuronal activity recorded was related to, at least, one of the 6 tasks, we used statistics to grade their activities with movements. The 50 bins were put into 6 sets in sequence from the first bin. The first and the last sets were composed of 9 bins and the remaining 4 were composed of 8 bins. Mean discharge rate in each of the 6 sets was calculated. Then the F-test was applied to detect whether all 6 mean values were equal or not. If the null hypothesis was rejected, the unit was considered to change its discharge rate in association with movement. Change of activity was graded as I ( + ) , II ( + + ) and III ( + + ÷ ) according to the significance levels of 1 ~o, 0.5 ~o and 0.1 respectively. A 1 ~o significance was adopted to contralateral movement, and 0.I ~o for ipsilateral movement. A severe criterion was held to ipsilateral movement in order to pick up a neuron that was surely related to it. The time of activity change corresponding to a movement was determined by eye on a histogram, where 3 successive bins increased (or decreased) sequentially.
EMG recordings Sometimes during the daily recording sessions, EMGs of muscles involved in forelimb movements were recorded in the identical situation to the unit study. A bipolar electrode, made of thin copper wires coated with enamel except for the tips, was inserted in various muscles accessible and identifiable from the surface. The recordings were repeated at least twice for each muscle on different occasions. In the case of a large muscle such as m. latissimus dorsi, EMGs were obtained from several different points. The upper part of this muscle was involved in all forelimb movements, while the lower part was almost inactive. Muscular activity was processed int o averaged histogram in a similar way to neuron activity. The amplitude of muscular activity was proportionally converted into the height in each bin of a histogram, using a comparator to discriminate a signal from noise. Forty-eight trials were averaged. Bin width was 10 msec for finger and wrist movements and 20 msec for arm movement. The following abbreviations were used for muscles; BC, m. biceps; BR, m. brachioradialis; DEX, deep extensor of the forearm; DFX, deep flexor of the forearm; DLT, m. deltoideus; ECR, m. ext. carpi radialis; FCR, m. flex. carpi radialis; ISP, m. infraspinatus; LAT, m. lattisimus dorsi; PCT, m. pectoralis major; RHM, m. rhomboideus; SCA, m. splenius capitis; SER, m. serratus ant.; SSP, m. supraspinatus; TC, m. triceps; TMA, m. terres major; TMI, m. terres minor; TRP, m. trapezius; 34F, genuine finger muscle between 3rd and 4th digits.
33 RESULTS
Recordings have been made from 197 neurons in the motor cortex of two monkeys. Sampling bias was carefully avoided for two points. First, neuronal activities were evenly recorded from the trunk and proximal limb area as well as from the distal limb area in the motor cortex. Second, any neuronal activity related to ipsilateral or contralateral movement was recorded. Out of 197 units, the discharge of 134 were associated with contralateral tasks, but not with any one of 3 ipsilateral tasks (contra-neurons). Sixty-three units discharged in relation to, at least, one of 3 ipsilateral tasks in addition to contralateral tasks. To these 63 neurons, the following criteria were applied to classify them into two groups. Criteria. (1) The peak and mean frequency during 50 bins around the onset o f movement were greater for ipsilateral than for contralateral movement and were greater than 10 spikes/sec.
A F
IPSI
CONTRA
tltllllll,i,hhl,hh,l,,I,l,,ll,,llhlll~,l~
w
,
,
(;
..
'
F
,
,,,,I,Itl.l,,,,....,..',,,,.l,.,,h.,.,h,,h~]
w
,
,
~
~
.
,',.~.___.
'" .ht,,,,,,,,h,ldldldDh~IHhhlllhlI
'"
i
I
llllilili,,,,,i,li,,ihl,,,l,ii,llillll OUTl ,,!,,,,.,,.,,~...,,,.i ', ,,dl,i,lhIlSli~l l
OUT
'
~
,
,
I B F
I PSI
hlLhl.
,. I L .
CONTRA
,
hi I l i l l l h l l l . h l l h l
,ill.
]
'.
I
.*'
, .
;"
]
..;
- ~ - - - I
W
i 0 I
IN ............,11.........!lllllll,!,,.,I....u ] .
.
.
.
.
IN
i
OUTi,,h,h,ll.,,.,,.,mL
,~ , , , I . h , I ,,., ,*,1 u
.
.
i, I
OUT
, o
Fig. 1. Example o f two ipsi-neurons in averaged time histograms. A a n d B: different neurons. The left column, labelled as IPSI, shows the histograms for ipsilateral finger (F), wrist (W), a r m m o v e m e n t in inward (IN) and outward (OUT) directions. The right column, labelled as C O N T R A , shows contralateral movements. A vertical bar represents 40 ° for arm displacement. A horizontal bar represents 200 msec for a r m and 100 msec for finger and wrist movements.
34
ant.
~
L f H s@E E(~S sH f f WSWW~S S B H f ~WWFwSWSF FWWWw(~B)WBB/~ .(~pfSBr .___s B e
--
~WW
E F(E)C S
~ . ~ w F W~)B
~B N N ~ a
Y
s.~"s N f
WwsSSSS
~eFFE FF S/'" F FWSF. I f W S W(t~,S W S~
E
~e°t
post.
Fig. 2. Representation of ipsi-neurons on the motor cortex. Figure shows two motor areas surrounded by the sulcus arcuatus (s.arc) anteriorly, by the sulcus centralis (s.cent) posteriorly, and by the sulcus precentralis superior (s.pre) medially. Each capital letter indicates the entrance of an electrode and the abbreviated name of a part of the body or forelimb. Encircledletters showthe locations ofipsi-neurons. Abbreviation; B, back; C, chest; E, elbow; F, finger; f, foot; H, hip; L, loin; 1,lip; N, neck; S, shoulder; W, wrist. (2) The unit activity started to increase 100 msec or earlier before the onset of ipsilateral movement. (3) The period from the onset of increase in unit discharge to the onset of the movement was longer for ipsilateral than for contralateral movement. Thirteen neurons completed all 3 criteria and they were defined as 'ipsi-neurons'. The other 50 did not complete all 3. They were defined as 'bilateral neurons'. In this report, description was restricted to the characteristics of the 13 ipsi-neurons. Fig. 1 shows examples of two ipsi-neurons in averaged time histograms. One neuron (Fig. 1A) changed its activity in all task performance on both sides. In the finger tasks (F), the activity decreased with the downward swing of the movements during ipsilateral and contralateral performances, but the change was more distinct in the ipsilateral movement. In the wrist tasks (W), the change was more remarkable in contralateral than in ipsilateral movement. For arm movements, the activity increased during ipsilateral inward movement (IN) and decreased during outward movement (OUT). This unit satisfied the 3 criteria for the ipsilateral inward movement. For the contralateral arm movement, the activity transiently increased then decreased during inward movement and gradually increased during outward movement. This neuron was fast P T N (0.7 msec antidromic latency) and ICS (47/~A) produced the contraction of shoulder muscles. In Fig. 1B, another ipsi-neuron is illustrated. It fired in all 3 ipsilateral movements and it was most distinct in the ipsilateral inward arm movement (IN). But, the unit was almost silent in all 3 contralateral tasks. This neuron was non-PTN and ICS (143 /,A) evoked the contraction of contralateral shoulder muscles. The response pattern of the remaining 11 ipsi-neurons is summarized in Table I with their other properties,
35
Location of ipsi-neurons in the motor cortex From histological data4,10,11, it could be inferred that ipsi-neuron activity could be more recorded in the trunk and proximal limb area in the motor cortex. To ascertain this inference, locations of ipsi-neurons were mapped on the motor cortex. In Fig. 2, capital letters denote the entrances of microelectrodes in a 1-mm grid on the motor cortex. The capital letters also represent the abbreviated names of parts of the body or the forelimb, identified by muscle contraction with ICS. The full name represented by each capital letter is given in the legend. The 12 encircled capital letters signify the tracks at which ipsi-neurons were encountered. Described in detail, 7 ipsineurons were obtained from 7 shoulder (S) tracks where ICS contracted shoulder muscles. Similarly, 1 ipsi-neuron was obtained from 1 back (B) track, 2 from one chest (C), 1 from 1 elbow (E), and 2 from 2 wrist (W) tracks. No ipsi-neurons were obtained from finger (F) tracks. Fig. 2 also shows that ICS contracted muscles in the lip (1), neck (N), hip (H), loin (L) and foot (f) in 17 tracks. But most neurons in these areas were not associated with any task performances. Six neurons were, however, related to one of the 6 tasks, one of them being bilateral neuron. The other 5, recorded in the hip, loin, or foot area, were contralateral units. Correlation of ipsi-neurons with muscular contractions To ascertain the foregoing inference further, the appearance rate of ipsineurons in the finger (F), wrist (W), elbow (E), shoulder (S) and trunk (T) tracks were separately studied. The chest and back were here included in the trunk. Six neurons related to muscular contraction of the lip, neck, loin, hip or foot were excluded. For the remaining 191 neurons, the following figures were obtained. Three ipsi-neurons (12.0~) were found in 25 T-neurons. Similarly, 7 ipsi-neurons (9.3 ~o) in 75 S-units, one (3.3 %) in 30 E-units, and 2 (6.3 %) in 32 W-units were also found. No ipsi-neurons were found in 29 F-units. Therefore, it was concluded that ipsi-neurons appeared with a higher rate in units related to trunk and proximal limb muscles. If T- and S-neurons were united into a 'proximal' group, and F-, W-, and E-neurons were categorized as a 'distal' group, the percentage of ipsi-neurons was 10 % (10/100) in the proximal and 3.3 % (3/91) in the distal group with a significant difference by z~-test (P < 0.05). Response pattern of ipsi-neurons to 6 movements Concerning the firing pattern of ipsi-neurons, it would be interesting to know the correlation of activities between finger, wrist or arm movement ipsilaterally, or the correlation between the ipsilateral and the contralateral activity. Table I summarizes this relation in 6 tasks. Five ipsi-neurons from the rostral motor cortex (RST-group) are shown in the upper part and 7 recorded near the sulcus precentralis superior (SPSgroup) are shown in the lower part of the Table. Four out of 5 RST-units responded to ipsilateral and contralateral arm movements, but neither to finger nor wrist movement. ICS contracted shoulder muscles in 3 of these 4 ipsi-units and back muscles in the remaining one ipsi-unit. One RST-unit was associated with ipsilateral and contralateral arm and finger movements, but the activity was more associated with ipsilateral movements. Thus RST ipsi-neurons were related to the arm rather than distal movements.
36
÷+÷÷ ÷÷~÷÷ ~÷÷÷÷ Z N ~
÷ ÷
~
÷÷~-~÷ ÷÷÷÷-~
It+++++ ÷ ÷
÷÷ ÷÷,÷÷
+l
I I I l l l
++++
÷÷÷
0
III1+
+
÷÷÷÷ ÷÷÷÷÷ • ÷÷÷÷ ÷÷÷÷÷
÷ ÷ +
÷
I I I I I +
÷ 0
I I I I+
÷
+
II
I+++
±÷÷~÷ -~-÷÷÷÷÷ ÷-~÷÷÷÷
I++++++
÷÷÷
÷÷÷
+++1+++ ÷÷ ÷÷
I++1+++
0
r~
r!. i
~
.~
o
÷ -~
~.
÷ ~-
37 O n the other hand, 5 o f the 7 SPS-units were associated with ipsilateral finger and wrist movements in addition to a r m movement. Three o f these 5 were also associated with contralateral finger, wrist and a r m movements. The fourth was not associated with any contralateral movement. The last one o f the 5 was related only to a r m m o v e m e n t contralaterally. But the activity was less significant than for the ipsilateral movement. O f the remaining two SPS-units, one was associated with ipsilateral a r m and with contralateral wrist and a r m movements. The other one was associated with only wrist movements on both sides. ICS produced wrist m o v e m e n t o f this neuron. The a m o u n t o f threshold current for ICS was different between SPS-group (252 -t- 132/zA) and R S T - g r o u p (542 4- 181 p A ) with statistical significance (P < 0.01). This reflected the fact that the a m o u n t o f the threshold current was smaller near the sulcus precentralis superior than in the rostral part o f the m o t o r cortex 20. Effects o f ICS were primarily contralateral to the stimulation. Ipsilateral muscle contractions were observed, but at that time contralateral contractions were as strong or dominant, and isolated ipsilateral muscular contraction was never observedS, e.
Ipsi-neurons and PTNs There are direct and indirect ipsilateral pathways to the spinal cord z0. F r o m this point, it would be interesting to k n o w the distribution o f ipsi-neurons in P T N and IPSI
CONTRA IA
IF
IW
IN
CA OUT
CF
34 F
CW '
~
IN ~ -
OUT
~lk
DEX OFX 8C
TC
BR
6
6
0
b
0
0
6
0
Fig. 3. EMGs of distal muscles. CONTRA: EMGs recorded from contralateral muscles during contralateral task performances. IPSI: EMGs recorded from contralateral muscles during ipsilatcral tasks. CF, CW, CA denote contralateral finger, wrist and arm movements. IF, IW, IA denote ipsilateral finger, wrist and arm movements. IN and OUT indicate inward and outward arm movements. BC, m. biceps; BR, m. brachioradialis; DEX, deep extensor muscle of the forearm; DFX, deep flexor of the forearm; ECR, m. ext. carpi radialis; FCR, m. flex. carpi radialis; TC, m. triceps; 34F, genuine finger muscles between 3rd and 4th fingers. Time scale, 250 msec for finger and wrist movements and 500 msec for arm movement before and after the onset of movements. A vertical bar on the right represents 100 pV.
38
IPSI IF
IW
CONTRA
IN
IA OUT
CF
CW
IN
CA
OUT
TRP
SER
I I
mmmm m
il
.,,, . - -
TM f
RHM
$sP
L SCA 0
0
0
0
0
O
o
o
Fig. 4. EMGs of proximal muscles. Explanation is the same as in Fig. 3. Abbreviation: DLT, m. deltoideus; IPS, m. infraspinatus; LAT, m. latissimus dorsi; PCT, m. pectoralis major; RHM, m. rhomboideus; SCA, m. splenius capitis; SER, m. serratus anterior; SSP, m. supraspinatus; TMA, m. terres major; TMI, m. terres minor; TRP, m. trapezius; Time scale, 250 msee for finger and wrist movements and 500 msee for arm movement before and after the onset of movements. Vertical bars represent 100 #V. non-PTN. The results were as follows; two (2.5 %) out of 80 PTNs and 11 (10.5 ~ ) out of 105 non-PTNs were ipsi-neurons. For the rest of neurons, the antidromic stimulation was not tested. These two ipsi-PTNs were fast-PTNs (Table I). The border latency to separate the fast and the slow was 2.0 msec zS. Moreover, ipsi-neurons were equally found in phasically as well as in tonically firing neurons, and in neurons with low spontaneous discharge rate as well as with higher frequency.
EMG during ipsilateral and contralateral tasks In consideration of ipsi-neurons, it is important to refer to muscular activities during ipsilateral and contralateral tasks. Fig. 3 a n d 4 are presented for this purpose. A time histogram of E M G was made in a similar way to unit study (cf. Methods). E M G s were recorded from the muscles contralateral to the recording cortex during the contralateral ( C O N T R A ) and ipsitateral (IPSI) task performances. Fig, 3 shows E M G s of 'distal' muscles. F o r all the muscles in this group, no significant activity was observed, when the monkey used the ipsilateral forelimb. Activity of biceps and triceps muscles was very weak during the task on either side. All other 'distal' muscles
39 changed their activities in, at least, one of 3 contralateral tasks, especially with the wrist. But it is necessary to comment on the EMG of finger muscles. When a EMG was recorded from the muscles between the 3rd and 4th fingers, as shown in Fig. 3, the activity was negligible during both ipsilateral and contralateral finger tasks. This reflected the performance of the monkey. During the contralateral finger task, the monkey depressed the small lever with index finger flexion, the 3rd and 4th fingers being kept in a loosely flexed state and out of use. When a EMG was recorded from the intrinsic finger muscles between the 2nd and 3rd digits, moderate activity was recorded during ipsilateral finger movement, though no activity was observed during the ipsilateral wrist movement. At that time, the monkey rested the contralateral hand quietly on a rod before its chest. Therefore, we could not suppress the activity of the contralateral finger muscles between the 2nd and 3rd digits during the ipsilateral finger task. In Fig. 4, the activity of trunk and 'proximal' muscles is illustrated. Most of the muscles in this group changed activity with contralateral finger (CF) and wrist (CW) movement. Changes of activities were also observed during the contralateral arm movements in the following muscles; mm. deltoideus (DLT), trapezius (TRP), serratus ant. (SER), terres major (TMA), pectoralis major (PCT) and infraspinatus (ISP). During ipsilateral task performances, activity change was detected in TRP, TMA, rhomboideus (RHM) and ISP muscles. Some muscles in this group were tonically active during the task with ipsilateral arm (IA). Activities of TRP, PCT and supraspinatus (SSP) had characteristics of 'distal' muscles in that they discharged phasically during CF and CW movements and were inactive during ipsilateral finger (IF) and wrist (IW) movements. Consideration of EMGs in relation to ipsi-neuron activity will be taken into account in the Discussion. DISCUSSION
Validity of criteria of ipsi-neurons The 3 criteria were adopted to define an ipsi-neuron (of. Results). These criteria seem to be severe, but still a question was raised. Are ipsi-neurons defined as such actually associated with the ipsilateral muscular activities, or is the change of activity in ipsilateral movement a secondary effect due to the contamination from contralateral movement? The discussion will be focused on this point in consideration with EMGs. For the 'distal' muscles, the contamination from contralateral movements can be neglected, because 'distal' limb muscles did not show any significant activities during ipsilateral performances. But in the case of the trunk and 'proximal' muscles, the result must be carefully interpreted. Some of the contralateral muscles in this group changed their activities in the ipsilateral tasks. As shown in Fig. 4, activities of the deltoideus (DLT), trapezius (TRP), terres major (TMA), rhomboideus (RHM), infraspinatus (ISP), and splenius capitis (SCA) muscles were modulated during ipsilateral arm movement. Activities during ipsilateral
40 tasks may possibly be due to the contamination from the contralateral movements. But this possibility can be excluded for the TRP and ISP muscles, because these muscles showed very significant change of activity during contralateral movements. If an ipsi-neuron was associated with these muscles, its activity was to be more related to contralateral than to ipsilateral movement. The SCA muscle can be excluded because it is a neck muscle and no ipsi-neuron activity was recorded from the neck area in the motor cortex. The DLT muscle was excluded because its activity was hardly modulated during ipsilateral task performances. But the TMA and R H M muscles could not safely be neglected only from the comparison of temporal pattern of EMGs and neuron activity. But the activities of these muscles during contralateral finger and wrist tasks were much more distinct than during ipsilateral tasks. Ipsi-neurons did not show such change of activity when they were associated with ipsilateral finger or wrist movement. Taken all together, it is more likely that ipsi-neurons were associated with ipsilateral rather than with contralateral movements.
Functional consideration of ipsi-neurons Based on the experimental and clinical data2-4,s,10,11, is, it was supposed that neuron activity associated with ipsilateral movement would be more related to trunk and proximal limb muscles. This supposition was confirmed by the present unit study. First, 12 out of the 13 ipsi-neurons were associated with ipsilateral arm movement and only one was solely related to wrist movement. Second, ICS provoked contraction of the trunk or shoulder muscles for 10 units and of elbow or wrist for the other 3 in their recorded locations. Third, the appearance rate of ipsi-neurons was higher in the proximal than in the distal group. In the present study, ipsi-units were mainly recorded in two different parts of the motor cortex; in the rostral (RST) motor cortex and near the sulcus precentralis superior (SPS). Units of two groups seem to have a little different function over ipsilateral movement. The RST-group was associated with ipsilateral arm movement, but the SPS-group was related to finger and/or wrist, in addition to arm movement. A somatotopical map 19 demonstrated that the RST-area corresponded to the trunk, and the SPS-area corresponded to the proximal limb muscles. Therefore RST-units would be responsible for ipsilateral regulation of the trunk or the control of the posture4,1°, while SPS-unit would be involved in ipsilateral proximal limb movement or limb-body integration4,1o. But the total number of ipsi-neurons was too small to attribute them the whole control of ipsilateral movement. Therefore, it seems more likely that ipsi- and bilateral neurons work in a cooperative manner to perform ipsilateral motor control, because the sum of ipsi- and bilateral neurons was 63 out of 197 movement-related neurons, constituting a fairly large number of the total. Comparison with the previous studies Three (7.5 %) out of 40 PTNs were associated with ipsilateral wrist extension 7. This figure was quite similar to the present result of 6.5 %. But if PTN and non-PTN were separately considered, a discrepancy arose. In the present experiment, 2 (2.5 %)
41 out of the 80 PTNs and 11 (10.5 ~ ) out of the 105 non-PTNs were ipsi-neurons. The ratio of ipsi-PTNs was very small in the present study. The discrepancy could be explained by the different criteria adopted in each report. The criterion previously reported 7 was the same as our first criterion, hence our criteria would be more severe than in the previous study, yielding a smaller ratio in the present report. The difference of the tasks might be another source of the discrepancy. Wrist extension was employed in the previous experiment 7, while we adopted 3 tasks using finger, wrist and arm. This procedure would enlarge the area from which ipsi-neuron activity was favoured to be recorded, by including the rostral motor cortex. Therefore, the difference of criteria may be the major reason for the discrepancy. Ipsi-neurons were significantly more numerous in n o n - P T N than in PTN. It is doubtful that the majority of PTNs were sampled from the distal limb area of the motor cortex, from which only a smaller portion of ipsilateral corticofugal fibers originateda,4,10,11. But this was to be excluded because 37 out of the 80 PTNs were sampled from the distal motor area TM.Probably very small number of ipsi-PTNs in the parent population would cause an exaggerated result. But still the fact may be taken as a demonstration of the importance of indirect ipsilateral pathways over the regulation of ipsilateral movement. ACKNOWLEDGEMENT The authors thank Prof. K. Kubota for his encouragement in the course of the experiment and Mrs. T. Miwa for her technical assistance and Miss S. Ishikara for the typing. REFERENCES 1 Asanuma, H. and Sakata, H., Functional organization of a cortical efferent system examined with focal depth stimulation in cats, J. NeurophysioL, 30 (1967) 35-54. 2 Brinkman, J. and Kuypers, H. G. J. M., Split brain monkey: cerebral control of contralateral and ipsilaterai ann, hand and finger movements, Science, 176 (1972) 536-539. 3 Brinkman, J. and Kuypers, H. G. J. M., Cerebral control of contralateral and ipsilateral ann, hand and finger movements in the split-brain rhesus monkey, Brain, 96 (1973) 653-674. 4 Brinkman, J., Split Brain Monkey: Cerebral ControlofContralateralandlpsilateralArm, Handand Finger Movements, Doctoral Thesis, Univ. of Rotterdam, 1974. 5 Bucy, P., Representation of ipsilateral extremities in the cerebral cortex, Science, 78 (1933) 418. 6 Bucy, P. and Fulton, J. F., Ipsilateral representation in the motor cortex of monkeys, Brain, 56 (1933) 318-342. 7 Evarts, E. V., Pyramidal tract neuron activity associated with a conditioned hand movement in the monkey, J. Neurophysiol., 29 (1966) 1011-1027. 8 Glee, P., Contra and ipsilateral motor and sensory representation in the cerebral cortex of monkey and man. In K. J. Ziilich, O. Creutzfeldt and G. C. Galbraith (Eds.), Cerebral Localization, Berlin, Heidelberg, New York, 1975, pp. 48-61. 9 Goldring, S. and Racheson, R., Human motor cortex: sensory input data from single neuron recordings, Science, 175 (1972) 1493-1495. 10 Kuypers, H. G. J. M., The descending pathways to the spinal cord, their anatomy and function. In J. C. Eccles and J. C. Schade (Eds.), Progress in Brain Research, VoL I 1, Organization of The Spinal Cord, 1964, pp. 188-202. 11 Kuypers, H. G. J. M. and Brinkman, J., Precentral projections of different parts of the spinal intermediate zone in the rhesus monkey, Brain Research, 24 (1970) 29-48.
42 12 Kwan, H. C., MacKay, W. A., Murphy, J. T. and Wong, Y. C., Spatial organization of precentral cortex in awake primates. II. Motor outputs, J. Neurophysiol., 41 (1978) 1120-1131. 13 Liu, C. N. and Chambers, W. W., An experimental study of the cortico-spinal system in the monkey (Macaea mulatta). The spinal pathways and preterminal distribution of degenerating fibers following discrete lesions, J. comp. Neurol., 123 (1964) 257-284. 14 Matsunami, K. and Hamada, I., Motor cortex unit activity related to ipsilateral upper limb movement of awake monkeys, XXVII int. Congr. PhysioL Sci. (Paris), (1977) 489. 15 Matsunami, K. and Hamada, I., Precentral neuronal activities associated with the upper limb movement in monkeys, J. Physiol. (Paris), 74 (1978) 319-322, 16 Matsunami, K. and Hamada, I., Antidromiclatency of the monkey pyramidal tract neurons related to ipsilateral hand movements, Neurosci. Lett., 16 (1980) 245-249. 17 Tigges, J., Nakagawa, S. and Tigges, M., Efferents of area 4 in a South American monkey (Saimiri). L Terminations in the spinal cord, Brain Research, 171 (1979) 1-10. 18 Twitehell, T. E., The restoration of motor function following hemiplegia in man, Brain, 74 (1951) 443-480. 19 Woolsey, C. N., Organization of somatic sensory and motor area of the cerebral cortex. In H. F. I-Iarlowand C. N. Woolsey ('Eds.), Biological and BiochemicaI Basis o f Behavior, Univ. of Wisconsin Press, Madison, Wise., 1958, pp, 63-81. 20 Woolsey, C. N., Settlage, P. H., Meyer, D. R., Senccr, W., Pinto, T. and Travis, A. M., Patterns of localization in precentral and 'supplementary' motor areas and their relation to the concept of a premotor area. Chap. XH. In Patterns of Organization in The Central Nervous System, Res. Publ. Ass. nerv. ment. Dis., 30 (1951) 238-264.