Dexterity in adult monkeys following early lesion of the ...tres/int2003/hall.pdf · Dexterity in adult monkeys following early lesion of the motor cortical ... was essentially for

European Journal of Neuroscience, Vol. 10, pp. 729–740, 1998 © European Neuroscience Association

Dexterity in adult monkeys following early lesion of themotor cortical hand area: the role of cortex adjacent tothe lesion

E. M. Rouiller, X. H. Yu, V. Moret, A. Tempini, M. Wiesendanger* and F. Liang†Institute of Physiology, Faculty of Sciences, University of Fribourg, Rue du Musee 5, CH-1700 Fribourg, Switzerland

Abstract

Infant monkeys were subjected to unilateral lesions of the motor cortex (mainly its hand representation). Aftermaturation, they showed normal use of the contralateral hand for global grip movements. However, as comparedwith the ipsilateral hand, precision grip tasks requiring relatively independent finger movements were performedwith less dexterity, particularly if adjustments of the wrist position were necessary. The purpose of this study wasto investigate mechanisms which may be responsible for the rather well, although not complete, preservation ofmanipulative behaviour of these adult monkeys. To this end, the hand representations were mapped bilaterallywith intracortical microstimulation in the mature monkeys, and the dexterity of both hands assessedquantitatively in a precision grip task. The behavioural effects of reversible inactivations of the primary (M1) andsupplementary (SMA) motor cortical areas were then tested. The following were found. (i) The hand contralateralto the lesion exhibited subtle but significant dexterity deficits, as compared with the ipsilateral hand; the deficitwas essentially for complex movements requiring dissociation of the thumb–index finger pinch from the otherdigits, involving also an arm rotation. (ii) Reversible inactivation of the M1 hand representation in the intacthemisphere dramatically impaired dexterity of the opposite hand without affecting the ipsilateral hand(contralateral to the early lesion). (iii) A relatively complete hand representation was found to occupy a newterritory, medial to the old lesion. (iv) The role of this new displaced representation was crucial for the preserveddexterity of the opposite hand, as evidenced by its functional inactivation. In contrast, inactivation of both SMAcortices did not interfere with the manipulative behaviour. It is thus concluded that the preserved functionalcapacity of manipulations with the hand opposite the early lesion can be essentially attributed to a corticalreorganization around the old lesion. Under the present experimental conditions, contributions from either theSMA or the intact M1 appear not to be crucial.

Introduction

As a consequence of unilateral lesion of the M1 hand representation,adult monkeys lose their digital skill of the opposite hand, such asthe precision grip (e.g. Denny-Brown & Botterell, 1947; Fulton,1949; Travis, 1955). The precision grip allows the prehension of asmall object between thumb and index finger. A similar devastatingeffect on prehension grip and wrist orientation for prehension hasbeen observed in adult monkeys when the hand representation in M1was reversibly inactivated (Rouilleret al., 1997; Kermadiet al., 1997;see also Fig. 5 in the present report). The precision grip movementof primates is believed to depend on monosynaptic connectionsbetween corticospinal neurones of the motor cortex and motoneuronesof the cervical spinal cord controlling the distal muscles of thehand (the ‘corticomotoneuronal system’—see, e.g. Kuypers, 1981;Lawrenceet al., 1985; Lemon, 1993a; Lawrence, 1994).

Functional recovery after motor cortical lesions, although usuallynot complete, is a typical and prominent feature as noted by many

Correspondence:Dr E. M. Rouiller, E-mail: [email protected]

*Present address: Department of Neurology, Laboratory of Motor Systems, University Hospital, Inselspital, CH-3010 Bern, Switzerland.

†Present address: Laboratory for Brain Structure and Function, Frontier Research Program, Institute of Physical and Chemical Research (Riken), 2–1 Hirosawa,Wako City, Saitama Pref. 351–01, Japan.

Received 11 February 1997, revised 9 July 1997, accepted 26 September 1997

authors (e.g. Travis, 1955), particularly with long survival times orwhen the motor cortex was lesioned shortly after birth (Kennard,1942). One of the principal deficit observed after lesions of the motorcortex or its pyramidal outflow is the loss of precision grip. It wasfound that the precision grip did not develop in monkeys subjectedto motor cortical (Passinghamet al., 1983) or pyramidal tract lesion(Lawrence & Hopkins, 1976). Casual (‘clinical’) observations ofmonkeys subjected to motor cortex lesions gave the impression thatthese animals developed normally to juvenile or adult age. To whatextent monkeys subjected to early lesion of M1 recover hand dexterityis still debated, as reflected by divergent claims for either full recovery(Kennard, 1942) or only incomplete functional restitution (Passinghamet al., 1983). The aim of the present work was therefore twofold:(i) to assess quantitatively the long-term effects on the precisiongrasp of monkeys that have been subjected to a unilateral motorcortex lesion as infants, and (ii) to establish potential compensatory

730 E. M. Rouilleret al.

TABLE 1. Summary of lesional and behavioural experiments

Behavioural scores*

Animals Lesion made at: Age at behavioural tests Vertical Horizontal

monkey 1 (S) 42 days 6 years NS P,0.01monkey 2 (A) 18 days 2 years NS P,0.01monkey 3 (P) 11 days 2 years NS NSmonkey 4 (N) 8 days 4 1/2 years NS P,0.01monkey 5 (B) 14 days 5 1/2 years P,0.01 P,0.01monkey 6 (M) 9 days 4 years P,0.01 P,0.01monkey 7 (B) 42 days 3 1/2 years P,0.05** P,0.01monkey 8 (C)*** no lesion 2 years NS NS

*Difference between the ‘normal’ and ‘affected’ hands (see text). **60 s test:P , 0.01; 45 s test:P , 0.05; 30 s test:P , 0.05. ***unlesioned animal, showingno significant difference (NS) between the two hands in the behavioural test.

mechanisms which may have developed during maturation to adult-hood. To reach these goals, we measured the behavioural performanceof the two hands, mapped the cortical hand representation in themotor cortex of both hemispheres with intracortical microstimulationsand used reversible inactivation techniques. It will be shown thatsubtle, but significant, deficits affecting some of the more refinedaspects of the performance of the precision grip did persist in thesemonkeys, such as a loss of the ability to perform the thumb–indexfinger pinch completely independently from the other fingers and adifficulty to combine synergically the precision grip with movementsof more proximal joints (e.g. rotation of the wrist, elevation of thearm). These data tend to support previous findings of incompletefunctional restitution (Lawrence & Hopkins, 1976; Passinghamet al.,1983). However, a new perilesional hand representation was formedthat was found to be crucial for the preservation of less elaborateaspects of hand dexterity, such as performance of the precision gripthat did not require complete independence with the other fingersand/or a synergy with other joints for proper orientation of the hand.

Materials and methods

Early infant lesion

Eight to 42 days after birth, the motor cortex of neonate macacafascicularis monkeys (n 5 7) was lesioned in the left hemisphereby extensive coagulation and subpial aspiration under ketamineanaesthesia (7.5 mg, initial dose). Three to four subsequent doses ofketamine (half of the initial dose) were given at 15–20-min intervals.The focus of the lesion was aimed at the hand area. The lesion cavitywas filled with gelatine sponge and the dura, fasciae and skin sutured.The infant monkeys were treated with penicillin and returned to themother in the colony in the animal room. For several years (Table 1),they grew up normally (clinging to the mother, sucking and laterjumping, climbing, etc.), together with unlesioned monkeys. Thelocation and extent of the lesions as they appear at the adult stageare shown in Figure 1.

Behavioural evaluation of manual dexterity

The general motor behaviour of the early lesioned animals in theircage appeared normal. However, to what extent were they capableof performing fine, independent, movements with the fingers of thehand deprived of corticospinal inputs? To address this question, theprecision grip ability of the (right) hand opposite the lesion wastested and compared with the performance of the left ‘normal’ hand.The behavioural evaluation was undertaken after maturation (2–6 years after the early lesion, see Table 1). We used the procedure

© 1998 European Neuroscience Association,European Journal of Neuroscience, 10, 729–740

described by Brinkman & Kuypers (1973). The animal was placedin a primate chair in front of a Perspex board (203 10 cm) containing50 rectangular slots (15 mm long, 6 mm wide and 6 mm deep), 25orientated vertically and 25 horizontally. Monkeys were trained during5–15 days to pick up a small piece of food in each slot, whichrequires the use of the precision grip, in alternating series with theleft or the right hand. Scoring started after the performance reacheda relatively stable level. Testing was repeated daily for 30 consecutivedays in order to obtain an average performance. In a typical session,the animal performed the task six times. In the first two tests (oneperformed with the left hand, the other with the right hand), themonkey was given for each test a period of 60 s to retrieve the foodmorsels from the 50 slots. The number of slots successfully retrievedwere counted. Later, after filling again all 50 slots with food, themonkey was given for each of the next two tests a period of 45 s,and scores were established. The same procedure was repeated forthe last two tests, except that in this case the test period was 30 s.Therefore, for each hand, three scores were established correspondingto test periods of 60, 45 and 30 s, respectively (see Figs 2–4).

Electrophysiological mappingAfter completion of the behavioural tests, under aseptic conditionsand pentobarbital anaesthesia (30 mg/kg), a chronic rectangular(330 mm by 250 mm) metallic chamber was implanted on the skull,and fixed with dental cement. The position of the chamber was suchthat the hand representation of M1 could be reached laterally on eachside of the chamber (Figs 7 and 8). The M1 somatotopic maps onthe intact and the lesioned side were established by standard proceduresfor intracortical microstimulation (ICMS, e.g. Asanuma & Rosen,1972; Sessle & Wiesendanger, 1982). Parameters of stimulation werethe same as those used in previous studies from this laboratory(Rouiller et al., 1994a,b, 1996): 35 ms duration trains of 12 electricpulses (0.2 ms), presented once every two seconds. Low impedance(0.1–0.3 MΩ) tungsten microelectrodes were advanced vertically fromthe pial surface along a distance of 10 mm, by steps of 0.5 mm. Ateach step, ICMS was performed, by applying initially a currentintensity of 50–90µA; then, the intensity was progressively decreasedto establish the threshold at which the corresponding movement ofthe contralateral body part was elicited. M1 on both sides was mappedextensively, along systematic rows of penetrations 1 mm apart fromeach other (Figs 7 and 8). About 10 electrode penetrations wereperformed on each daily ICMS session, during which the monkeyreceived food as reinforcement.

Transient inactivationAs will be described in the Results, ICMS revealed a displaced handrepresentation in the lesioned M1 cortex. In order to test its function,

Neonatal lesion of M1 in monkeys 731

FIG. 1. For each early lesioned monkey,drawing of a frontal Nissl-stained sectionof the left hemisphere indicating thelocation and extent of the lesion (shadedarea), as it appears at the adult stage. Thesection was taken at a rostrocaudal levelcorresponding to the maximal extent of thelesion. The arrow points to the centralsulcus.

this cortical area was transiently inactivated by microinfusions of thelocal anaesthetic lidocaine. Such experiments were performed on twomonkeys (Table 2), participating in an inactivation session once aweek, four times in monkey 7 and eight times in monkey 5. On thatday, the animal first performed the behavioural task (control scores)and a cortical area (see below) was reversibly inactivated by infusinglidocaine (4% solution in saline), as previously described in detail(Martin, 1991; Rouiller et al., 1997). Martin (1991) establishedthe time course and the spatial extent (drug spread) of reversibleinactivation, and provided evidence in control experiments that similarinjections of saline had no effect. Lidocaine was injected by pressureusing a 10µL Hamilton microsyringe. After 5–10 min postinfusion,the behavioural task was repeated to assess the effect of inactivation.Thirty to 60 min later, when the action of lidocaine faded out, thebehavioural task was repeated a third time to check that the animalrecovered from the transient inactivation. In an inactivation session,


lidocaine was injected in one of the following areas: hand area ofM1 right (M1-R), hand area of M1 left (M1-L), SMA right (SMA-R), SMA left (SMA-L). Twelve inactivation sessions were conductedon monkeys 5 and 7 as listed in Table 2, providing information onthe cortical area inactivated, the volume of lidocaine injected and thenumber of infusion sites. In two sessions conducted on monkey 5(Table 2), SMA-R was inactivated first, followed immediately byinactivation of SMA-L, to test the effect of the simultaneous inactiva-tion of both SMAs.

To infuse lidocaine in M1-R or in M1-L, the microsyringe wasorientated vertically (as the electrode tracks for ICMS) and placedon the pial surface exactly at the same coordinates as the correspondingelectrode penetrations along which ICMS elicited distal movements(thumb, fingers or wrist). Penetrations were usually not perpendicularto the surface and thus traversed a number of representations ofdifferent fingers, wrist or elbow. To avoid infusion of lidocaine in


FIG. 2. Typical behavioural scores obtained in monkeys 6 and 5 in theprecision grip task, comparing the performance of the left (‘normal’) and theright (‘affected’) hands. The early lesion was aimed at the left M1. Scorescorrespond to the number of slots successfully retrieved during the three testperiods considered (60, 45 and 30 s). Scores are average values with SDs of30 daily sessions. In each daily session, the three test periods were performedby the animal one after the other in that order. Scores are given separatelyfor the vertical slots (top panels) and the horizontal slots (bottom panels). Theleft hand exhibited higher scores than the right hand; the difference wasstatistically significant (P , 0.01; Mann–Whitney test).

more proximal representations than that of the wrist, the depth of themicrosyringe with respect to the pial surface was also matched to thedepth of the sites at which distal ICMS effects were obtained at lowcurrents (generally, 10 µA). With the precise matching of ICMSand microsyringe penetrations, it is reasonable to assume that lidocaineinfusions were limited to the hand representation. Mixed distal-proximal representations were not selected as targets for lidocaineinfusion in M1. Altogether in monkeys 5 and 7, three and fourinactivations were performed in M1-R and M1-L, respectively(Table 2).

In the SMA, location of the arm representation was roughlyestablished bilaterally based on elbow or shoulder movements elicitedby ICMS, to determine the appropriate mediolateral coordinate.Two distinct inactivation sessions were performed on monkey 7 toinactivate SMA unilaterally (Table 2): in the first one (session 4),lidocaine (1.5µL) was injected in SMA-R 4 mm below the pialsurface in each of four penetrations along the rostrocaudal axis,covering SMA-proper; in the second SMA experiment (inactivationsession 5), lidocaine (same volume) was infused in correspondingpoints in SMA-L (four penetrations along the rostrocaudal axis).Injection at sites located rostrally with respect to the elbow or shoulderrepresentations indicate that the inactivation also involved the handarea, according to the somototopy of SMA (Macphersonet al.,1982; Wiesendanger, 1986; Mitz & Wise, 1987). In monkey 5, twoexperiments (Table 2: sessions 6 and 7) were performed to inactivateSMA bilaterally. In both SMA-R and SMA-L, lidocaine was injectedin four penetrations at roughly the same locations as in monkey 7and in three additional penetrations located more rostrally to includepre-SMA. In each penetration 1.0µL was injected at each of twosites, 5 and 3 mm below the pial surface, respectively. Finally, stillin monkey 5, lidocaine was infused in SMA and pre-SMA unilaterally,on the right side (Table 2: inactivation session 8).


Results

Lesions

The location and extent of the lesions as they appear at the adultstage were established on the basis of Nissl-stained sections. For eachearly lesioned monkey, a drawing of a frontal section taken in themiddle of the rostrocaudal extent of the lesion is shown in Figure 1.In monkeys 1, 2 and 4–7, the lesion was placed in the precentralgyrus, covering a part of the rostral wall of the central sulcus as wellas a part of the above crest region of the central sulcus, thuscorresponding to the representation of the hand and the arm in M1.In monkey 3 in contrast, the lesion was restricted to the postcentralgyrus and therefore did not include M1, in line with the behaviouraldata (see below). Typically, the lesion appeared either as a missingpart of the cerebral cortex or as a damaged cortical zone free ofthe typical large pyramidal neurones of the intact M1 hand/armrepresentation. The reconstructions in Figure 1 show that all lesionsare incomplete and therefore do not cover the entire extent of area 4at the level of the hand representation. For instance, in monkey 2,the lesion does not include most of the depth of the central sulcus.

Behavioural evaluation of manual dexterity

To assess possible normal hand asymmetries in the Brinkman boardtask, an additional intact animal (monkey 8) was tested and wasfound to have a non-significant difference in performance for the twohands (Table 1, Mann–Whitney test,P . 0.5). Moreover, the scoresobtained in the lesioned monkeys using their ‘normal’ hand weresimilar to the scores of the intact monkey (Mann–Whitney test,P . 0.05).

To pick up the food morsel out of the slots with the ‘normal’ hand,the monkeys typically first inserted the index finger in the verticalslot and then brought the thumb into opposition, as previouslydescribed in detail (Brinkman, 1984). The task was more difficult forthe horizontal than the vertical slots because, for successful picking, themonkey had to make the thumb–index finger pinch more independentlyfrom the other digits in the horizontal than in the vertical slots.Furthermore, grasping in the horizontal slots was more complexbecause the monkey had first to rotate the wrist for proper orientation(synergy of rotation and prehension). The performance was assessedduring 30 daily sessions, by counting the number of objects success-fully retrieved and brought to the mouth during periods of 60, 45and 30 s.

Behavioural scores obtained for two representative lesionedmonkeys are shown in Figure 2. Monkey 6 showed a significantpreservation of precision grip ability with the hand opposite the lesionfor the vertical slots only; however, the scores were significantlylower than the scores obtained for the normal hand (Table 1; Mannand Whitney test,P , 0.01). Retrieval of the objects from thehorizontal slots was only possible for monkey 6 when using thenormal hand (Fig. 2). In contrast, monkey 5 was also able to performthe precision grip with the hand opposite the lesion in both the verticaland horizontal slots (Fig. 2). However, for both slot orientations, thescores of the hand opposite the lesion were significantly lower thanthe scores obtained for the normal hand (Table 1; Mann–Whitneytest, P , 0.01). Among the other five lesioned monkeys (Table 1),three animals (monkeys 1, 2 and 4) exhibited scores for the handopposite the lesion that were not significantly lower than the scoresobtained for the ‘normal’ hand in the vertical slots. In contrast, thescores in the horizontal slots were significantly lower for the handopposite the lesion. One animal (monkey 7, see Table 1) showedbehavioural data comparable with those obtained for monkey 5(Fig. 2). Finally, the last animal (monkey 3) presented similar scores


FIG. 3. Relationship between daily sessionnumber (time in days) and behaviouralscores for monkey 4, given for verticalslots (left panels) and horizontal slots (rightpanels). The performance of the left(‘normal’) hand and the right hand(opposite the lesion) in the precision griptask is indicated by open circles and filledsquares, respectively. In panels A and D,the test period was 60 s, while it was 45 sin panels B and E, and 30 s in panels Cand F.

for both hands (Table 1), but it turned out that the lesion was locatedin the postcentral gyrus and did not encroach the precentral (M1)hand representation (Fig. 1). Differences among the behaviouralcapacities of the various lesioned monkeys (e.g. deficits restricted toretrieval from horizontal slots or affecting also retrieval from verticalslots) were not related to the postnatal age at which the lesion wasmade (ranging from 8 to 42 days postnatal: see Table 1).

To summarize the behavioural data (average performance, seeFig. 2 and Table 1), the hand contralateral to the early lesion presenteda deficit of skilful manipulation, observable mainly for the mostchallenging horizontal slots. Of the six monkeys (monkey 3 havingbeen eliminated), three exhibited a deficit restricted to the synergy ofthe combined hand orientation with the precision grip (horizontalslots), while the precision grip itself (vertical slots) was normal.However, as mentioned above, precision grip may be in itself moredifficult for the horizontal slots, requiring more independent finger


movements than for the vertical slots. In the other three monkeys,there was, in addition to the deficit for the horizontal slots, a deficitin performance for the vertical slots also, with fewer trials, butrelatively normal appearance of the precision grip.

Although animals were trained for 5–15 days before starting formalassessment, some lesioned monkeys continued to improve their scoresover the 30 days of behavioural evaluation. Thus, monkey 4 exhibiteda progressive improvement for vertical slots (Fig. 3A–C), but thiswas also the case for the control hand. A comparable training effectwas observed in the three test periods (60, 45 and 30 s; Fig. 3A–C,respectively). For horizontal slots (Fig. 3D–F), a training effect waspresent for the control hand while the score of the hand opposite thelesion remained nearly zero (no successful picking); the improvementof the performance of the normal hand for horizontal slots was moreprominent in the 60 s test period than in shorter test periods (Fig. 3D–F). In sharp contrast to monkey 4 (Fig. 3), monkey 7 did not


FIG. 4. Relationship between daily sessionnumber (time in days) and behaviouralscores for monkey 7, given for verticalslots (left panels) and horizontal slots (rightpanels). Same conventions as in Fig. 3.Note the absence of training effect in thisanimal.

show clear practice effects (Fig. 4). In between monkeys 4 and 7characterized either by a training effect for both vertical and horizontalslots (Fig. 3) or no training effect at all (Fig. 4), respectively, othermonkeys exhibited a training effect for one or the other of the twoslot orientations. Monkey 5 showed progressively increasing scoreswith both hands for the horizontal slots but not for the vertical ones.This was the reverse in monkey 6, exhibiting an effect of trainingfor the vertical slots, similar to that observed for monkey 4 (Fig. 3A–C), but no score increase for the horizontal slots with the normal hand.

Possible mechanisms of plasticity

One can hypothesize that some preservation of manual skills (essen-tially grasping from the vertical slots), in monkeys subjected to earlylesions of the opposite M1-hand representation, was due to plasticchanges of the brain resulting from the lesion. Possible mechanismsof such plasticity were investigated using two approaches: transient


reversible inactivation of a restricted brain area (by microinfusion ofthe local anaesthetic lidocaine) and electrophysiology (intracorticalmicrostimulation). The following three hypothetic mechanisms weretested:

Hypothesis 1: contribution of M1 on the intact hemisphere

From comparable experiments conducted in rats, there was behaviouraland neuroanatomical evidence that the intact opposite hemispherecan contribute to functional restitution by redirecting a fraction ofcorticospinal (CS) axons to the denervated cervical cord (Barth &Stanfield, 1990; Rouilleret al., 1991). To test this hypothesis inour monkeys, the intact M1 hand representation was transientlyinactivated. In monkeys 5 and 7, three sessions of transient inactivationof the intact M1-R were performed (Table 2), by infusing the localanaesthetic lidocaine (40µg/µL saline, volume ranging from 8 to18 µL distributed across six to nine sites, covering the extent of the


TABLE 2. Summary of the characteristics of the inactivation sessions performed on monkeys 5 and 7

Number of sites ofAnimal Cortical area inactivated Total volume of lidocaine infused per area infusion

Session 1 Monkey 7 M1-R 10µL 6Session 2 Monkey 5 M1-R 18µL 9Session 3 Monkey 5 M1-R 8µL 6Session 4 Monkey 7 SMA-R* 6µL 4Session 5 Monkey 7 SMA-L* 6µL 4Session 6 Monkey 5 SMA1pre-SMA 14µL 14

bilaterallySession 7 Monkey 5 SMA1pre-SMA 14µL 14

bilaterallySession 8 Monkey 5 SMA-R1pre-SMA-R 14µL 14Session 9 Monkey 7 M1-L† 5µL 4Session 10 Monkey 5 M1-L† 8µL 5Session 11 Monkey 5 M1-L† 10µL 10Session 12 Monkey 5 M1-L† 18µL 9

*Infusion of lidocaine restricted to the caudal part of SMA, corresponding to SMA-proper.†Infusion of lidocaine performed in the territory adjacent to the lesion (see text).Session 7 was a repetition of session 6, 1 week later. The results were the same (see text).

hand area characterized by ICMS). A typical result of such aninactivation experiment in M1-R (session 2, see Table 2) is illustratedin Fig. 5 (left column). When comparing the performance in theprecision grip task before and after inactivation, it turned out that thebehavioural score of the hand opposite the lesion was not modified,while the performance of the normal hand was dramatically reducedas expected. A second inactivation session of M1-R was performedon the same animal, in addition to a third inactivation session of M1-R conducted on monkey 7 (Table 2). In both cases, the results werecomparable with those illustrated in Figure 5 (left column), i.e. noeffect on the hand contralateral to the lesion and a dramatic deficitof the normal hand, ipsilateral to the lesion. However, the latterdeficit was a bit less pronounced in the inactivation sessions 1 and 3as compared with session 2 (Fig. 5, left column), as one could expectfrom the infusion of smaller volumes of lidocaine (Table 2). Theseinactivation data call for a rejection of this hypothesis; in other words,M1 in the intact hemisphere does not appear to play a significantpart in the incomplete preservation of skilful manipulation followingearly lesion of the hand representation of the opposite M1, at leastin the present experimental conditions.

Hypothesis 2: contribution of the supplementary motor area

As non-primary motor areas also contain corticospinal neurones (e.g.Dum & Strick, 1991), one can speculate that such areas (for instance,the premotor cortex5 PM or the supplementary motor area5 SMA)can substitute for the lesioned M1 cortex. In particular, there isevidence, at light microscopic level, that the CS axons originatingfrom the SMA may give rise to corticomotoneuronal connectionswith cervical motoneurones controlling distal hand muscles (Rouilleret al., 1996). To test the idea that the SMA may contribute to somepreservation of skilful manipulation after early lesion, we performedreversible inactivations of the SMA in monkeys 5 and 7, eitherunilaterally (SMA left or SMA right) or of both SMAs simultaneously(Table 2). In the three sessions of inactivation of the SMA conductedin monkey 5, lidocaine was infused in each SMA at seven locationsalong the same rostrocaudal line, 2 mm apart. At each location, 1µLof the lidocaine solution in saline was injected at two different depths(sites) from the pial surface (2 and 4 mm, respectively). The totalvolume injected in each SMA was then 14µL. The arm/hand areawas roughly located on the basis of ICMS. However, the seven sites


FIG. 5. Effect of transient reversible inactivation of M1 (right or left) on thebehavioural scores (precision grip task, 60 s) obtained in two representativesessions on a lesioned animal (monkey 5). Inactivation was induced byinfusion with a Hamilton microsyringe of a lidocaine solution (40µg/µLsaline). Lidocaine was injected in M1-R (left column) or in M1-L (rightcolumn) at locations previously determined electrophysiologically(intracortical microstimulation, ICMS) to correspond to fingers and wristrepresentations. Sites injected were distant from each other by 2–3 mm inorder to cover the hand representation entirely. In the right M1 (M1-R: leftcolumn), lidocaine was infused in the intact hand representation while in theleft M1 (M1-L: right column), lidocaine was infused in the ‘re-mapped’territory adjacent to the lesion (see text), where distal movements were elicitedby ICMS. Inactivation in M1-R and M1-L corresponds to the sessions 2 and12, respectively (Table 2). Scores were established before injection of lidocaine(control score: two leftmost bins in the plots), 10–20 min after injection(action of lidocaine: two middle bins in the plots) and 60–70 min afterinjection (recovery due to fading of lidocaine action: two rightmost bins inthe plots). The behavioural scores for this monkey (5) were lower than inFig. 2 because in that condition the head was fixed by the implanted chamberand extensive ICMS mapping produced small lesion of the cortical tissuewhich might have reduced the performance.


FIG. 6. Effect of transient reversible inactivation of the SMA (by infusion oflidocaine) on the behavioural scores (precision grip task; 60 s) obtained in alesioned animal (monkey 5). The right SMA (SMA-R) was inactivated first(scores in bins 3 and 4 from the left in the plots), immediately followed byinactivation of the left SMA (SMA-L), then corresponding at that stage to abilateral SMA inactivation (scores in bins 5 and 6 from the left). The rightmosttwo bins are scores after recovery (60–70 min after infusion in the left SMA).

of injection extended on more caudal and rostral territories, in orderto obtain a broad inactivation of the whole SMA, including the SMA-proper and the pre-SMA (see Wiesendanger, 1986; Matsuzakaet al.,1992; Luppinoet al., 1993 for the distinction between a rostral anda caudal zone in the SMA). Typical results of a bilateral inactivationof SMA are illustrated in Fig. 6. In this experiment (Table 2, session6), lidocaine was infused first in SMA-R (14µL at 14 sites) andbehavioural scores were established. Immediately afterwards,lidocaine was then infused in SMA-L (14µL at 14 sites) followedby another behavioural evaluation. In the latter test, one can considerthat both SMAs were inactivated simultaneously. The behaviouralscores after SMA-R inactivation and after inactivation on both sideswere unchanged, both for the hand opposite the lesion and the normalhand (Fig. 6). The same experiment was repeated in monkey 5(Table 2: session 7), and comparable results with those presented inFigure 6 were obtained. Still in monkey 5, a third session wasperformed in which only SMA-R was inactivated (14µL in the same14 sites as above). Here again, there was no change in the behaviouralscores established before and after inactivation. In monkey 7, twoinactivation sessions of the SMA were performed (Table 2). In session4, only SMA-R was inactivated while in session 5 it was SMA-L. Inboth sessions, lidocaine was injected at four sites along the rostrocau-dal axis (2 mm apart from each other), representing a total volume


of 6 µL in each session. In that case, the injections were aimed tothe caudal region of the SMA (SMA-proper) and most likely did notinclude the pre-SMA. Once more, there was no effect on thebehavioural scores of either hand, confirming the data derived fromthe other animal. Thus, the inactivation data derived from theexperiments conducted on monkeys 5 and 7 indicate that the SMAneither seems to play an important part in the precision grip task (forthe normal hand) nor in the incomplete preservation of skilledmotricity by the hand opposite to the early lesion.

Hypothesis 3: contribution of cortical territory adjacent to thelesion

Some preservation of skilled hand and finger movements may dependon the reorganization of cortical territories adjacent to the lesion.This possibility was investigated by comparing maps obtained byICMS in the two hemispheres. In a group of intact monkeys (n 5 4),ICMS experiments clearly demonstrated that the hand representationsin M1 on the two hemispheres were symmetrically located, occupyingsimilar rostrocaudal and mediolateral stereotaxic coordinates (notshown). In an early lesioned animal (monkey 2), the hand representa-tion in the intact M1 (right hemisphere) was located in a zonecorresponding to that of normal monkeys, extending from rostrocaudalcoordinates 14–19 mm and mediolateral coordinates 13–16 mm(Fig. 7). On the left hemisphere, the lesion was identified on thecorresponding histological sections, i.e. in a zone normally corres-ponding to the hand representation (Fig. 7). However, ICMS per-formed in territories medial to the lesion, normally devoted tomore proximal muscles (elbow, shoulder, back, leg), elicited distalmovements of the hand opposite the lesion, such as fingers and wristmovements (Fig. 7). As a consequence, proximal representations wereshifted more medially indicating that, in spite of the lesion, asomatotopic representation of the whole forelimb was preserved, butshifted by 3–4 mm in the medial direction in the case of the monkeyillustrated in Figure 7.

Further evidence for such reorganization of a hand area of thecortical territory adjacent to the lesion was investigated in similarextensive ICMS experiments conducted on four additional animalssubjected to an early lesion of M1 hand representation (monkeys 3,5, 6 and 7). A comparable map reorganization in the territory adjacentto the lesion is illustrated in Figure 8 for monkey 6. On the lefthemisphere, ICMS performed at the location of the presumed handrepresentation had no effect (triangles in Fig. 8), indicating that thisterritory has indeed been lesioned. Medially to the lesioned area, ina zone normally devoted to the representation of proximal muscles(elbow, shoulder and back), ICMS was found to elicit distal movements(thumb and wrist). Comparing this to the other (intact) hemisphere,it appears that this reshaped hand representation has been moved notonly medially, but also rostrally (Fig. 8). Remapping of the territoryadjacent to the lesion as shown in Figures 7 and 8 was observed alsoin monkeys 5 and 7 (not shown). The lesion in monkey 3 was not inthe M1 hand representation (Fig. 1) and, consistently, the handrepresentation established by ICMS occupied a position correspondingto that found in the other hemisphere or in control monkeys. TheseICMS data in monkeys 2 and 5–7 provide evidence that the corticalterritory adjacent to the lesion has been remapped.

The contribution of this remapped territory to some preservationof skilled manual performance was supported by a dramatic decreasein the behavioural score of the hand opposite the lesion when this ‘new’(shifted) hand area was transiently inactivated pharmacologically(Table 2: sessions 9–12). A typical result of such an inactivation ofM1-L in the territory adjacent to the lesion is illustrated in Figure 5(right column, session 12). As expected, inactivation of this remapped


FIG. 7. Somatotopic map established in M1by intracortical microstimulation (ICMS)on both hemispheres in a representative(lesioned) animal (monkey 2). Therectangle corresponds to the borders of themetallic chamber chronically implanted onthe monkey’s skull after completion of thebehavioural tests. The vertical line atcoordinate 0 mm represents the mid-line.Rostral is on top, left hemisphere is on theleft. Coordinates are given in millimetres.Location of the lesion is indicated byshaded area. Triangle and circle symbolsindicate location of electrode penetrationsfor ICMS. The size of the circles indicatesthe threshold inµA (table on top of thefigure) at which a movement was elicitedby ICMS. The corresponding body partactivated was indicated by the letters(according to conventions given on the topof the figure). The territory correspondingto the hand representation (T, D and Wpoints) is outlined on each hemisphere.Note a shift of the hand representationmedially in the left (lesioned) hemisphereas compared with the intact (right)hemisphere (see text).

FIG. 8. Somatotopic map established in M1by intracortical microstimulation (ICMS)on both hemispheres in a representative(lesioned) animal (monkey 6). Sameconventions as in Figure 7. Triangle andcircle symbols indicate electrodepenetrations for ICMS. The location of thelesion in the M1-L area corresponds to thezone where ICMS had no effects(triangles), covering approximately themediolateral coordinates 10–15 mm androstrocaudally 14–19 mm. Shaded areacorresponds to zone with scar tissue. Thehand representation (T, D and W points) isoutlined on each hemisphere. Note medialshift of the hand representation in the left(lesioned) hemisphere as compared withthe intact (right) hemisphere (see text).

territory did not affect the behavioural performance of the normalhand, ipsilateral to the microinfusion of lidocaine (Fig. 5, rightcolumn). On the contrary, the hand contralateral to the lesion, afterlidocaine infusion, completely lost the ability to retrieve food morselsfrom the vertical slots, as compared with preinfusion data (Fig. 5,


right column). In this experiment of M1-L inactivation, lidocaine wasinjected in nine sites in the remapped area representing a total volumeof 18 µL. Comparable results as in Figure 5 (right column) wereobtained in three other inactivation experiments of M1-L conductedon monkeys 5 and 7 (Table 2: sessions 9, 19 and 11), although the


deficit of the hand contralateral to the lesion was not as dramatic asin session 12, in line with the smaller volumes of lidocaine infusedin the sessions 9–11 (Table 2).

Summary of the data

In conclusion, the above reversible inactivation and ICMS dataindicate that, in our experimental conditions, M1 in the intacthemisphere and the SMA on either side did not take over control ofthe hand function contralaterally to the early cortical lesion. Theobservation that the intact M1 does not appear to play a crucial partis consistent with the absence of clear lesion-related changes of theefferent projections from the intact M1 after removal of the oppositesensorimotor cortex in infant monkeys (Sloperet al., 1983). Thepresent ICMS and reversible inactivation data support the idea thatplastic modifications of the somatotopic representation in the lesionedhemisphere contribute significantly to some preservation of skilledhand–finger movements following the early motor cortex lesion ofthe hand representation.

Discussion

Partial preservation of manual dexterity after infant lesion ofM1

Even so the adult monkeys that were subjected to an early lesion ofM1 appeared to move normally on general observation, a detailedassessment of manual dexterity revealed that the preservation ofmanual skill was incomplete (Fig. 2; Table 1), because from the sixmonkeys in which the lesion affected the M1 hand representation,two animals completely failed (score5 0) to perform the precisiongrip in the horizontal slots, requiring a synergy of finger and wristmovements, in agreement with the observations of Passinghamet al.(1983). In these two animals, the deficit was mainly a failure to adaptthe wrist in the optimal orientation when performing the precisiongrip and to execute the grips with the thumb and index fingerindependently from the other fingers. When there was some preserva-tion of precision grip ability for the horizontal slots as well (e.g.monkey 5 in Fig. 2), it turned out that the extent of the cortical lesionwas smaller than in the monkeys which were successful for thevertical slots only (e.g. monkey 6 in Fig. 2).

Even for vertical slots, skilful hand manipulation was not completelynormal, because scores of the hand opposite the lesion were signific-antly lower than of the normal hand (Fig. 2). For the vertical slots,the discrepancy between the study of Passinghamet al. (1983) andthe present study might reflect differences in the size of the lesionand in criteria to consider a trial as successful. The lesion performedby Passingham and his coauthors were larger than our lesions.According to Passinghamet al. (1983), animals were ‘cheating’ whenthey picked up the food with the affected hand from the vertical slots(‘pulling and clawing the object rather than picking it out’). Ouranimals clearly picked the object between thumb and index finger ofthe affected hand, but more slowly than with the normal hand, asindicated by the smaller success rate during the same amount of time.In addition, the grip movement of thumb and index finger of thehand opposite the lesion was less independent of the other fingers ascompared with the normal hand, sufficiently independent however,to avoid interference of the other fingers with the neighbouring slots.Possible explanations for the persisting deficit may include, amongother factors, the smaller size of the remapped hand territory ascompared with that on the intact hemisphere and/or the diminishednumber of corticospinal axons originating from the newly formedhand area, adjacent to the lesion. It remained that the hand opposite


the lesion brought the thumb into opposition with the index finger topick up the object out of the slot, which was the only way for themonkey to obtain relatively good scores for vertical slots (Fig. 2).Importantly, the retention of some manipulatory skills was demon-strated by the dramatic decrease in the performance of the handopposite the lesion after reversible inactivation of the ‘reorganized’territory adjacent to the lesion (Fig. 5).

Examination of the lesions in Figure 1 clearly shows that only partof area 4 was affected. For instance, in monkeys 2 and 5, the lesionwas limited to the convexity of the central sulcus while, in monkeys1, 4, 6 and 7, it encroached a part of both the convexity and depthof the sulcus. Previous studies provided evidence for subdivisionswithin the arm representation of the monkey motor cortex (see, e.g.Lemon, 1981; Tanji & Wise, 1981; Strick & Preston, 1982; Stepniew-skaet al., 1993); recently, in human (Geyeret al., 1996), an area 4p(in the depth of the sulcus) was distinguished from an area 4a (onthe convexity). It may well be that lesion of area 4a does not affecthand manipulative skills in the same way as after the lesion of 4p.This issue needs to be addressed in future experiments.

Plastic mechanisms: comparison with previous studies

The present results in the monkey clearly differ from conclusionsmet in rats subjected to unilateral neonatal lesions of M1, in whichCS and corticothalamic axons originating from the intact hemispherewere redirected (Barth & Stanfield, 1990; Rouilleret al., 1991; Yuet al., 1995). This discrepancy can be explained by developmentaldifferences between the two species. In rats, at the time of the lesion(postnatal days 0–2), the CS projection has not yet reached targetsin the cervical cord, leaving open the possibility to redirect CS axonsto a denervated territory. In contrast, in infant monkeys at a timecorresponding to that of the lesions performed in the present set ofexperiments, CS projections to the intermediate zone at the C8/T1spinal level were clearly present (Armandet al., 1994). Therefore, asthe CS axons have already invaded the spinal grey matter, a redirectionto the opposite denervated side is less likely to occur than in the rat.Although in neonatal monkeys the CS axons are already present inthe intermediate zone of the cervical cord, it should be emphasizedthat the CS projection is not yet functional; it will mature progressivelyover a period of 11 months, and this is paralleled by the developmentof skilled finger movements (Armandet al., 1994). At birth, themonkey brain (representing 40–50% of the volume of the adult brain)is clearly more mature than the rat brain, representing at birth onlyabout 10% of the volume of the adult brain (Passinghamet al., 1983).

A reorganization of the somatotopic map in the cortical territoryadjacent to the lesion was observed after M1 lesions in juvenile (2–2.5 years old) Rhesus monkeys (Glees & Cole, 1950). These authorsperformed small lesions of the thumb representation (as revealed bycortical surface stimulation) and observed that ‘adjacent areas of theremaining and still intact parts of the motor cortex develop a latentability to control the functions of the previously altered areas’.Subsequent undercutting of this perilesion territory caused a reappear-ance of the previously observed deficit. The consistency of the presentlong-term observations in monkeys lesioned at infancy with those ofGlees & Cole (1950) lesioned at later stages (half grown animals)indicates that comparable plastic mechanisms may take place. Interes-tingly, a reorganization of the cortical territory adjacent to the lesionas observed in monkeys by Glees & Cole (1950) and in the presentstudy, has been found in rats too, when bilateral M1 lesions weremade at adult age (Castro-Alamancos & Borrell, 1995). It can beconcluded that, in rats, both the stage of development at the time ofthe lesion as well as the extent of the lesion may play a major part


in triggering different plastic changes contributing to preservationand/or recovery of motor function.

The present experiments do not answer the question whether theage of the monkeys when they incurred the lesions was playing apart in preserving some manual skills, or whether it was simply thelong survival time. It is known that, even in mature monkeys, recoveryafter motor cortical lesions is remarkable (e.g. Travis, 1955). Ameaningful comparison of the power of plasticity of the immaturelesioned brain with that of the adult lesioned brain will only be possiblewhen survival times (and training?) are comparable. Therefore, thevalidity of the Kennard principle (Kennard, 1942) remains to betested in future experiments with comparable lesions and survivaltimes in neonatal and adult animals.

Post-lesion plasticity of representational maps in the motor cortexin the adult depends strongly on use of the affected hand, asdemonstrated in two recent reports on the effects of focal ischaemiclesions affecting a part of the M1 hand representation in the squirrelmonkey (Nudo & Milliken, 1996; Nudoet al., 1996b). It was observedthat focal ischaemia induced a reduction in the distal representationaround the lesion (corresponding to the well-known distant effect ordiaschisis) but an apparent increase of proximal representations. Theauthors concluded that in monkeys, not subjected to postlesion motortraining, the movements previously represented in the lesioned corticalterritory do not reappear in zones adjacent to the lesion (Nudo &Milliken, 1996). However, if after focal ischaemic lesion, the monkeyswere subjected to ‘rehabilitative’ training for skilled hand use, thisprevented the loss of the hand representation adjacent to the lesionand, in some cases, an expansion of the hand representation in regionspreviously devoted to proximal movements was observed (Nudoet al., 1996b).

In the present study, the new cortical hand representation, adjacentto the lesion, was shifted medially, corresponding to territories initiallydevoted to proximal forelimb and trunk (maybe leg) movements. Thequestion whether the new hand representation was also extendingaround the lesion in the lateral direction could not be answeredbecause the implanted chamber did not extend far enough laterallyto access the facial representation. We did also not explore possiblechanges taking place in the somatosensory cortex. It has indeed beenshown that the motor function of the somatosensory cortex in monkeyscan contribute to compensate for the deficits due to cooling of themotor cortex (Sasaki & Gemba, 1984). However, one has to keepin mind that reversible inactivation and permanent lesions do notalways produce the same results (Lomberet al., 1996).

We suggest that the demonstrated lesion-related representationalplasticity of the motor cortex underlies the relatively well preservedskilled hand and finger movements after early lesion. Extensivereorganization of cortical maps has been described in a large numberof reports in both sensory and motor systems (see for review, e.g.Merzenichet al., 1988; Kaas, 1991; Ponset al., 1991; Garraghty &Kaas, 1992). Concerning more specifically the motor cortex of ratsand monkeys, reshaping of the somatotopic map was also observedin relation to peripheral nerve lesion (Saneset al., 1988; Donoghue& Sanes, 1988), to prolonged stimulation (Nudoet al., 1990), toblockade of inhibitory intracortical connections (Jacobs & Donoghue,1991), and also was found to be activity dependent (Nudoet al.,1996a). Comparable plastic changes of the somatotopic map wereobserved with functional imaging in humans, either in relation toincreased motor skill learning (Karniet al., 1995) or as a consequenceof a cortical lesion (Chollet & Weiller, 1994). In the latter case, PETdata indeed showed activation of new areas adjacent to the lesion, aswas the case in the present study. In addition, in adult patients witha unilateral brain lesion, plastic changes were not limited to the


hemisphere homolateral to the lesion, but involved recruitment ofcortical areas also in the undamaged hemisphere (Lemon, 1993b;Chollet & Weiller, 1994). Under the present experimental conditions,ICMS mapping failed to demonstrate plastic changes of the somato-topic map in the intact hemisphere. This discrepancy may reflect amajor difference in the timing of the lesion, performed at the infantstage in our monkeys while, in the human subjects, functional imagingdata are based on lesions that took place at the adult stage. Inmonkeys, evidence for a part possibly played in functional recoveryof motricity by ipsilateral corticospinal fibres has been found (Kucera& Wiesendanger, 1985); however, the experimental conditions weredifferent from the present study because the unilateral lesion wasperformed at the level of the bulbar pyramid and at the adult stage.Another parameter to consider is the size of the lesion. One cannotexclude that, in our monkeys, a larger lesion of M1 would have ledto a reorganization of the intact hemisphere as well. A perilesionreorganization may well take place only if some tissue of the M1arm area is left intact. It is conceivable that other mechanisms suchas vicarious functioning of other motor areas or somatosensory cortex,or the motor cortex of the intact hemisphere may come into play,which could not be demonstrated in the present experiments. Futureexperiments are needed to extend the observations of Glees & Cole(1950), by performing larger lesions, at a later stage of developmentand using the most modern tracing methods to investigate the plasticchanges. A number of studies in monkeys (e.g. Travis, 1955) and inneurological patients (e.g. Foerster, 1936) have described the remark-able recovery, sometimes including manual dexterity, after motorcortical lesions. Finally, monkey experiments will also enable us totest to what extent such postlesion functional recovery at the adultstage can be improved by pharmacological interventions or bymanipulating environmental factors (see for review Johansson &Grabowski, 1994; Johansson, 1995).

Acknowledgements

The authors thank Dr V. Maier and Dr B. Hyland for their contribution toearly steps of the project. We thank A. Schwartz, J. Corpataux and B. Morandifor taking care of the monkeys in the animal room. The present work wassupported by the Swiss National Science Foundation (grants number 3130–025128, 31–28572.90, 31–43422.95 to EMR and grants number 31–27569.89,31–36183.92 and 4038–044053 to M.W.), the Roche Research Foundation(Basle, Switzerland) and the CIBA-GEIGY Jubila¨ums-Stiftung (Basle, Switzer-land). We thank Mr Robert Scheiber for his editorial assistance.

References

Armand, J., Edgley, S.A., Lemon, R.N. & Olivier, E. (1994) Protractedpostnatal development of corticospinal projections from the primary motorcortex to hand motoneurones in the macaque monkey.Exp. Brain Res.,101, 178–82.

Asanuma, H. & Rose´n, I. (1972) Topographical organization of corticalefferent zones projecting to distal forelimb muscles in the monkey.Exp.Brain Res., 14, 243–56.

Barth, T.M. & Stanfield, B.B. (1990) The recovery of forelimb-placingbehavior in rats with neonatal unilateral cortical damage involves theremaining hemisphere.J. Neurosci., 10, 3449–59.

Brinkman, C. (1984) Supplementary motor area of the monkey’s cerebralcortex: short- and long-term deficits after unilateral ablation and the effectsof subsequent callosal section.J. Neurosci., 4, 918–29.

Brinkman, J. & Kuypers, H.G.J.M. (1973) Cerebral control. of contralateraland ipsilateral arm, hand and finger movements in the split-brain rhesusmonkey.Brain, 96, 653–74.

Castro-Alamancos, M.A. & Borrell, J. (1995) Functional recovery of forelimbresponse capacity after forelimb primary motor cortex damage in the rat isdue to the reorganization of adjacent areas of cortex.Neuroscience, 68,793–805.


Chollet, F. & Weiller, C. (1994) Imaging recovery of function following braininjury. Curr. Opin. Neurobiol., 4, 226–30.

Denny-Brown, D. & Botterell, E.H. (1947) The motor function of the agranularfrontal cortex.Res. Publ. Ass. Nerv. Ment. Dis., 27, 235–345.

Donoghue, J.P. & Sanes, J.N. (1988) Organization of adult motor cortexrepresentation patterns following neonatal forelimb nerve injury in rats.J. Neurosci., 8, 3221–32.

Dum, R.P. & Strick, P.L. (1991) The origin of corticospinal projections fromthe premotor areas in the frontal lobe.J. Neurosci., 11, 667–89.

Foerster, O. (1936) Motorische Felder und Bahnen. In Bumke, O. & Foerster, O.(eds),Handbuch der Neurologie, Band 6, Springer-Verlag, Berlin, pp. 1–357.

Fulton, J.F. (1949)Physiology of the Nervous System, 3rd edn. OxfordUniversity Press, New York, p. 667.

Garraghty, P.E. & Kaas, J.H. (1992) Dynamic features of sensory and motormaps. Curr. Opin. Neurobiol., 2, 522–7.

Geyer, S., Ledberg, A., Schleicher, A., Kinomura, S., Schormann, T., Bu¨rgel,U., Klingberg, T., Larsson, J., Zilles, K. & Roland, P.E. (1996) Two differentareas with the primary motor cortex of man.Nature, 382, 805–7.

Glees, P. & Cole, J. (1950) Recovery of skilled motor functions after smallrepeated lesions of motor cortex in macaque.J. Neurophysiol., 13, 137–48.

Jacobs, K.M. & Donoghue, J.P. (1991) Reshaping the cortical motor map byunmasking latent intracortical connections.Science, 251, 944–7.

Johansson, B.B. (1995) Functional recovery after brain infarction. A reviewof experimental animal data.Cerebrovasc.Dis., 5, 278–81.

Johansson, B.B. & Grabowski, M. (1994) Functional recovery after braininfarction: Plasticity and neural transplantation.Brain Pathol., 4, 85–95.

Kaas, J.H. (1991) Plasticity of sensory and motor maps in adult mammals.Annu. Rev. Neurosci., 14, 137–67.

Karni, A., Meyer, G., Jezzard, P., Adams, M.M., Turner, R. & Ungerleider,L.G. (1995) Functional MRI evidence for adult motor cortex plasticityduring motor skill learning.Nature, 377, 155–8.

Kennard, M.A. (1942) Cortical reorganization of motor function.Arch. Neurol.Psychiatry, 48, 227–40.

Kermadi, I., Liu, A., Tempini, A., Wiesendanger, M. & Rouiller, E.M.(1997) Differential effects of reversible inactivation of the primary motor,supplementary motor, dorsal premotor, and posterior parietal cortices, onreaching, grasping, and bimanual movements.Somatosensory and MotorRes., 14, 268–280.

Kucera, P. & Wiesendanger, M. (1985) Do ipsilateral corticospinal fibersparticipate in the functional recovery following unilateral pyramidal lesionsin monkeys?Brain Res., 348, 297–303.

Kuypers, H.G.J.M. (1981) Anatomy of descending pathways. In Brooks, V.B.(ed.), Handbook of Physiology(The Nervous System), Vol. II, Part 1.American Physiological Socety, Bethesda, MD, pp. 597–666.

Lawrence, D.G. (1994) Central neural mechanisms of prehension.Can.J. Physiol. Pharmacol., 72, 580–2.

Lawrence, D.G. & Hopkins, D.A. (1976) The development of motor control.in the Rhesus monkey: evidence concerning the role of corticomotoneuronalconnections.Brain, 99, 235–54.

Lawrence, D.G., Porter, R. & Redman, S.J. (1985) Corticomotoneuronalsynapses in the monkey: light microscopic localization upon motoneuronsof intrinsic muscles of the hand.J. Comp. Neurol., 232, 499–510.

Lemon, R.N. (1981) Variety of functional organization within the monkeymotor cortex. J. Physiol. (London), 311, 521–40.

Lemon, R.N. (1993a) Cortical control. of the primate hand.Exp. Physiol., 78,263–301.

Lemon, R.N. (1993b) Cortical plasticity: Stroke recovery.Curr. Biol., 3, 463–5.Lomber, S.G., Payne, B.R. & Cornwell, P. (1996) Learning and recall of form

discriminations during reversible cooling deactivation of ventral-posteriorsuprasylvian cortex in the cat.Proc. Natl Acad. Sci. USA, 93, 1654–8.

Luppino, G., Matelli, M., Camarda, R. & Rizzolatti, G. (1993) Corticocorticalconnections of area F3 (SMA-proper) and area F6 (pre-SMA) in themacaque monkey.J. Comp. Neurol., 338, 114–40.

Macpherson, J.M., Marangoz, C., Miles, T.S. & Wiesendanger, M. (1982)Microstimulation of the supplementary motor area (SMA) in the awakemonkey.Exp. Brain Res., 45, 410–6.

Martin, J.H. (1991) Autoradiographic estimation of the extent of reversibleinactivation produced by microinjection of lidocaine and muscimol in therat. Neurosci. Lett., 127, 160–4.

Matsuzaka, Y., Aizawa, H. & Tanji, J. (1992) A motor area rostral to the


supplementary motor area (presupplementary motor area) in the monkey:Neuronal activity during a learned motor task.J. Neurophysiol., 68, 653–62.

Merzenich, M.M., Recanzone, G., Jenkins, W.M., Allard, T.T. & Nudo,R.J. (1988) Cortical representational plasticity. In Rakic, P. & Singer, W.(eds),Neurobiology of Neocortex. John Wiley and Sons Ltd, Chichester.

Mitz, A.R. & Wise, S.P. (1987) The somatotopic organization of thesupplementary motor area: Intracortical microstimulation mapping.J. Neurosci., 7, 1010–21.

Nudo, R.J. & Milliken, G. (1996) Reorganization of movement representationsin primary motor cortex following focal ischemic infarcts in adult squirrelmonkeys.J. Neurophysiol., 75, 2144–9.

Nudo, R.J., Jenkins, W.M. & Merzenich, M.M. (1990) Repetitivemicrostimulation alters the cortical representation of movements in adultrats.Somatosens. Motor Res., 7, 463–83.

Nudo, R.J., Milliken, G.W., Jenkins, W.M. & Merzenich, M.M. (1996a) Use-dependent alterations of movement representations in primary motor cortexof adult squirrel monkeys.J. Neurosci., 16, 785–807.

Nudo, R.J., Wise, B.M., SiFuentes, F. & Milliken, G.W. (1996b) Neuralsubstrates for the effects of rehabilitative training on motor recovery afterischemic infarct.Science, 271, 1791–4.

Passingham, R.E., Perry, V.H. & Wilkinson, F. (1983) The long-term effectsof removal of sensorimotor cortex in infant and adult rhesus monkeys.Brain, 106, 675–705.

Pons, T.M., Garraghty, P.E., Ommaya, A.K., Kaas, J.H., Taub, E. & Mishkin,M. (1991) Massive cortical reorganization after sensory deafferentation inadult macaques.Science, 252, 1857–60.

Rouiller, E.M., Liang, F., Moret, V. & Wiesendanger, M. (1991) Trajectoryof redirected corticospinal axons after unilateral lesion of the sensorimotorcortex in neonatal rat; A phaseolus vulgaris-leucoagglutinin (PHA-L) tracingstudy.Exp. Neurol., 114, 53–65.

Rouiller, E.M., Babalian, A., Kazennikov, O., Moret, V., Yu, X.-H. &Wiesendanger, M. (1994a) Transcallosal connections of the distal forelimbrepresentations of the primary and supplementary motor cortical areas inmacaque monkeys.Exp. Brain Res., 102, 227–43.

Rouiller, E.M., Liang, F., Babalian, A., Moret, V. & Wiesendanger, M. (1994b)Cerebellothalamocortical and pallidothalamocortical projections to theprimary and supplementary motor cortical areas: A multiple tracing studyin macaque monkeys.J. Comp. Neurol., 345, 185–213.

Rouiller, E.M., Moret, V., Tanne´, J. & Boussaoud, D. (1996) Does the handrepresentation of the supplementary motor cortical area make direct contactswith cervical motoneurons?Eur. J. Neurosci., 8, 1055–9.

Rouiller, E.M., Yu, X.H. & Tempini, A. (1997) Effect of inactivation of M1and SMA on precision grip performance in monkeys. In Hepp-Reymond,M.C. & Marini, G. (eds),Perspectives of Motor Behavior and its NeuralBasis, Karger, Basel, pp. 33–43.

Sanes, J.N., Suner, S., Lando, J.F. & Donoghue, J.P. (1988) Rapidreorganization of adult rat motor cortex somatic representation patternsafter motor nerve injury.Proc. Natl Acad. Sci. USA, 85, 2003–7.

Sasaki, K. & Gemba, H. (1984) Compensatory motor function of thesomatosensory cortex for the motor cortex temporarily impaired by coolingin the monkey.Exp. Brain Res., 55, 60–8.

Sessle, B.J. & Wiesendanger, M. (1982) Structural and functional definitionof the motor cortex in the monkey (macaca fascicularis).J. Physiol.(London), 323, 245–65.

Sloper, J.J., Brodal, P. & Powell, T.P.S. (1983) An anatomical study of theeffects of unilateral removal of sensorimotor cortex in infant monkeys onthe subcortical projections of the contralateral sensorimotor cortex.Brain,106, 707–16.

Stepniewska, I., Preuss, T.M. & Kaas, J.H. (1993) Architectonics, somatotopicorganization, and ipsilateral cortical connections of the primary motor area(M1) of owl monkeys. J. Comp. Neurol., 330, 238–71.

Strick, P. & Preston, J.B. (1982) Two representations of the hand in area 4 ofa primate. I. Motor output organization.J. Neurophysiol., 48, 139–49.

Tanji, J. & Wise, S.P. (1981) Submodality distribution in sensorimotor cortexof the unanesthetized monkey.J. Neurophysiol., 45, 467–81.

Travis, A.M. (1955) Neurological deficiencies after ablation of the precentralmotor area in Macaca mulatta.Brain, 78, 155–73.

Wiesendanger, M. (1986) Recent developments in studies of the supplementarymotor area of primates.Rev. Physiol. Biochem. Pharmacol., 103, 1–59.

Yu, X.-H., Moret, V. & Rouiller, E.M. (1995) Re-examination of the plasticityof the corticothalamic projection after unilateral neonatal lesion of thesensorimotor cortex in the rat: a phaseolus vulgaris-leucoagglutinin tracingstudy.J. Brain Res., 36, 123–33.

Documents

Dexterity in adult monkeys following early lesion of the ...tres/int2003/hall.pdf · Dexterity in adult monkeys following early lesion of the motor cortical ... was essentially for