Persistence of vibrissal motor representation following vibrissal pad deafferentation in adult rats

Transcription

1 Exp Brain Res (2001) 137: DOI /s RESEARCH ARTICLE Gianfranco Franchi Persistence of vibrissal motor representation following vibrissal pad deafferentation in adult rats Received: 21 April 2000 / Accepted: 6 November 2000 / Published online: 16 February 2001 Springer-Verlag 2001 Abstract The effect of sensory vibrissal pad denervation on M1 organization was studied in adult rats 2 weeks after the infraorbital nerve was severed. Cortical motor output organization was assessed mapping the representation size and thresholds of vibrissa movements evoked by intracortical electrical microstimulation (ICMS). Motor cortex output patterns of control and sham groups of rats were compared with those of rats that had received unilateral or bilateral infraorbital nerve lesions. The mean size of the vibrissa representation in both unilateral and bilateral input-deprived hemispheres was not significantly different from those in control and sham hemispheres. The mean threshold required to evoke vibrissa movements was significantly higher in both groups of deafferented hemispheres than in control and sham groups of hemispheres. In contrast, the mean threshold required to evoke other types of movements from both groups of input-deprived hemispheres were similar to those found in the control and sham groups of hemispheres. These results indicate that input-deprived vibrissal motor representation reflects lower-than-normal excitability, although the size and topographic relationship with neighboring representations are normal. Keywords Motor cortex Trigeminal deafferentation Cortical reorganization Plasticity Introduction Neurophysiological studies show that, after peripheral nerve lesion, input-deprived sensory cortical areas and output-deprived motor cortical areas are occupied by neighboring representational fields (for review, see Sanes and Donoghue 1997; Kaas 1999). Sensory input to G. Franchi ( ) Dipartimento di Scienze Biomediche e Terapie Avanzate, Sezione di Fisiologia umana, Università di Ferrara, Ferrara, Italy fhg@dns.unife.it Tel.: , Fax: the sensory cortex is essential for the maintenance of sensory maps as motor output is essential for the maintenance of motor maps. In the motor cortex of adult animals, the organization of movement representation can be largely modified by manipulations of peripheral motor nerves (Sanes et al. 1988; Donoghue et al. 1990). In adult rats, M1 output reorganization stabilizes during the second week following contralateral facial nerve lesion. After this period, forelimb and eye representations firmly occupy neighboring vibrissa representations (Sanes et al. 1990; Toldi et al. 1996). Following the reinnervation of the facial nerve, a re-representation of contralateral vibrissae movement reorganizes in a shrunken region localized in the medial part of the former vibrissa representation. The remaining portion of the vibrissa representation was firmly occupied by forelimb and eye representations, showing no signs of the reversal of the expansion displayed before facial nerve reinnervation (Franchi 2000). In contrast, the role of sensory input in maintaining representational maps in the motor cortex of adult animals is still unclear. In the rat, thalamic nuclei and somatosensory cortical areas relay peripheral sensory input to the M1 (Miyashita et al. 1994; Izraeli et al, 1995; Porter 1996), thus affecting the physiological properties of the MI neurons (Sievert et al. 1986; Kaneko et al. 1994). In adult rats, it has been shown that sensory feedback from the forelimb plays a role in short-term reshaping of motor cortical output. A sustained increase in the degree of input effectiveness to MI, caused by stretching forelimb muscles, expands the forelimb representation into the vibrissal representation (Sanes et al. 1992). In neonatal rats it has been shown that vibrissal sensory input plays a significant role in shaping representational maps in the motor cortex (Keller et al 1996; Huntley 1997b). Vibrissa trimming from birth induces shrinkage of the vibrissa movement representation that persists after trimming has ceased. In adult rats, the motor cortex was resistant to the effects of unilateral vibrissa trimming (Huntley 1997b) or the effects of bilateral vibrissa trimming reversed 5 days after clipping ceased (Keller et al 1996).

2 The reversion of the shrinkage in the whisker movement representation presumably reflects the post-trimming resumption of normal whisking behavior. The present study was undertaken to define the role sensory input plays in maintaining representational maps in the motor cortex of adult animals. To this purpose, change in the size and excitability of motor representation was examined after persistent ablation of peripheral sensory input. Since most peripheral nerve manipulations involve damage to both sensory and motor axons, the decision was made to test the effect lesioning of a purely sensory nerve would have on MI output organization. In these experiments, the infraorbital nerve (ION) the nerve supplying sensory innervation to vibrissal and muzzle skin was used because it is anatomically separate from the vibrissal motor innervation and because the mystacial vibrissae have a relatively large representation in the rat MI cortex (Donoghue and Wise 1982; Neafsey et al. 1986). In order to study how tonic vibrissa input ablation affects MI organization, the effect of vibrissal pad deafferentation on MI excitability and representation patterns was investigated 2 weeks after infraorbital nerve lesion. Previous study evidenced that 2 weeks of recovery provides ample time for short-term modulatory influences on motor cortex reorganization (Sanes et al. 1988, 1990; Donoghue et al. 1990; Toldi et al. 1996) so that after this period detectable changes can be ascribed to long-term modulatory influences of somatic sensory input on motor cortical output. Since a large percentage of neurons in the somatosensory cortex receive bilateral vibrissal input (Armstrong-James and George 1988) the effect of unilateral and bilateral vibrissal deafferentation was tested in two groups of animals. These results were compared with those obtained in control and sham animals. Materials and methods Experiments were carried out on 28 male Albino Wistar rats, weighing g, divided into four groups of seven animals. In two groups of experimental rats, the infraorbital nerve (ION) was severed at the point where it exits from the infraorbital foramen: the surgery was unilateral in one group (Unilateral deafferented group), and bilateral in the other (Bilateral deafferented group). One of the two remaining groups was left untouched (Control group) and the last group underwent surgery on one side: the infraorbital nerve was isolated from surrounding tissues but left intact (Sham group). Infraorbital nerve surgery and follow-up 181 All surgical procedures were performed under ketamine anesthesia (100 mg/kg IP and then supplemental dose IM as needed). Using aseptic procedures, under the operating microscope, the infraorbital nerve was exposed, separated from its adjacent tissues and legated; then it was cut distally to eliminate all remaining fine branches. The proximal stump was dried and covered with acrylic tissue adhesive (Histoacryl) to prevent the proximal axons from sprouting. The skin was closed with 6-0 sutures and the wound was cleansed with an antibiotic solution. In the post-operative period unilaterally and bilaterally injured animals, like the controls and shame groups, displayed bilateral rhythmical vibrissa movements during natural whisking as they explored freely in their cages. During natural whisking the deafferented vibrissa did not suddenly retract when it hit against targets, as would normally be the case. After deafferentation the vibrissal pad proved unreactive to light pain-inducing sensory stimuli (i.e. light touch, squeeze or piercing). The loss of vibrissal pad sensitivity following deafferentation was clearly evidenced in all animals for the entire survival period. Intracortical stimulation mapping In each animal ICMS-evoked movements in the frontal agranular cortex were mapped. In the rats that underwent surgery, cortical mapping was carried out after the surgical wound had healed (14 18 days after surgery). Mapping was aimed at defining the extent of the vibrissa representation and the current threshold required to elicit vibrissa movement in the hemisphere contralateral to the side that underwent surgery. The mapping procedure was similar to the one described by Donoghue and Wise (1982) and Sanes et al. (1990), and detailed elsewhere (Franchi 2000). Briefly, the animals were placed in a Kopf stereotaxic apparatus and the frontal cortex of one side was exposed by a large craniotomy. The dura remained intact, and was kept moist with a 0.9% saline solution. The electrode penetrations were regularly spaced out over a 500 µm grid. Alteration in the coordinate grid, up to 200 µm, was sometimes necessary to prevent the electrode from penetrating the surface blood vessels. These adjustments in the coordinate grid were not reported in the reconstructing maps and were taken into account for quantitative map construction only when greater than 50 µm. Glass insulated tungsten electrodes (0.6 1 MΩ impedance at 1 khz) were used for stimulation. The electrode was lowered vertically to 1.5 mm below the cortical surface and adjusted to ±200 µm so as to evoke movement at the lowest threshold. In a previous experiment this depth was found to correspond to layer V of the frontal agranular cortex (Franchi 2000). Cathodal monophasic pulses (30 ms train duration at 300 Hz, 200 µs pulse duration) of a maximum of 60 µa were passed through the electrode with a minimum interval of 2.5 s. Starting with a current of 60 µa, intensity was decreased in 5 µa steps until the movement was no longer evoked; then intensity was increased to a level at which nearly 50% of the stimulations elicited movement. This level defined the current threshold. If no movements or twitches were evoked with 60 µa, the site was recorded as negative. When movement was observed in two different body parts or bilateral whiskers, current thresholds were determined for each component. In general, at threshold current levels, only one movement was elicited from any given point. Body parts activated by microstimulation were identified by visual inspection and/or muscle palpation. During the experiment, when at rest, forelimbs and hindlimbs were approximately half-way between flexion and extension, and were alternately flexed and extended, particularly when defining representational borders. Electrophysiologic facial nerve testing At the conclusion of the mapping session, the compound muscle action potential (MAP) of vibrissa muscles was recorded (Fig. 1A). The recording and stimulating procedure was similar to the one described by Archibald et al. (1991) and detailed elsewhere (Franchi 2000). All MAPs were obtained by supramaximal nerve stimulation, and recorded by computer data acquisition for off-line analysis (Spike2 for Windows). MAP latency was calculated using cursors in 20 responses for each recording side. The mean MAP latency in control animals was similar to those found in animals that had undergone surgery (injured: 1.5±0.09 ms versus control: 1.6±0.23 ms; P>0.1; t-test, P<0.05). The area under the curve of the averaged, rectified MAP provides a better estimate of the number of muscle fibers than does the peak-to-peak amplitude (Archibald et al 1991). The integration of MAP waveforms showed no significant differences between the injured animals and

3 182 In each animal, gross post-perfusion examination of the injured ION evidenced no continuity of the nerve at the level of the acrylic stopper. Under the operating microscope, care was taken to control that all ION fascicles had been tied and axotomized. In some animals, a few axons had sprouted at the periphery of the acrylic stopper without clear reinnervation of distal vibrissal pad. In all the animals studied, the exposed ION was cut and prepared for histological examination proximal to the site of nerve injury. After postfixation in osmium tetroxide, toluidine blue was used to stain 1 µm thick sections. Morphological examination of sections (Axioskop Zeiss and DMC Polaroid camera for image acquisition) showed extensive degenerative processes involving all axons proximally to the lesion (Fig. 1B). Map construction Fig. 1 A Evoked motor action potential (MAP) recorded in control and deafferented vibrissa. Both MAPs present similar amplitude, latency and duration. MAP was recorded by two monopolar needle electrodes inserted subcutaneously at symmetrical points in the vibrissal array of the two sides. Two monopolar needles, inserted percutaneously just posterior to the ear, were used as the facial nerve stimulating electrodes. The reference electrode was another needle inserted midway between the stimulating and the recording electrodes. The facial nerve was then stimulated with a rectangular current pulse of 0.2 ms duration. The position of the stimulating electrodes was optimized by obtaining a threshold of MAP at 1 ma. All MAPs were obtained by supramaximal nerve stimulation. B Light photomicrograph cross section of infraorbital nerve proximal to the acrylic stopper 2 weeks postinjury ( 100 magnification, osmium tetroxide and toluidine blue stain). Note evidence of Wallerian degeneration and/or perineural thickening in all ION nerve fascicles (insert: 400 magnification) the control animals (injured: 4.8±1.13 mv.ms versus control: 5.4±2.38 mv.ms; P>0.2; t-test level of significance, P<0.05). This observation suggests that the surgical lesion did not damage the mystacial muscle pad and its motor innervation. Histology At the end of the experimental procedure, the animals were perfused transcardially. The brains were sectioned and stained with thionine to verify microelectrode positions and depths. Penetration grid maps were constructed by labeling electrode penetrations according to the distance (in mm) from the bregma and by plotting threshold movements at a current intensity of 60 µa or less. Map borders were defined as the mid-point between sites with different movement thresholds. If a movement site was separated by a site where no movement was observed up to 60 µa (non-effective site), the borderline for the represented movement was placed 250 µm from the movement site. In each hemisphere, vibrissa movements were mapped in order to determine the size and location of the vibrissa representation. To reduce the effect of experimental sources of variability, similar mapping density was maintained across all animals. The resulting sizes of individual vibrissa representations in the hemispheres were compared using unpaired Student s t-test (level of significance, P<0.05). The hindlimb and forelimb movements were elicited around a whisker area; nevertheless the representations of these movements were not mapped in sufficient detail to determine size. Moreover, the cortex medial to the vibrissa representation was not systematically explored to define in detail its excitability and the extension of representations of meiosis or eye-eyelid movements. To evaluate the mediolateral reorganization of vibrissa representation, the percentage laterally situated sites of vibrissa movements was determined (Sanes et al. 1990; Huntley 1997b). To this purpose, penetrations spaced over a 0.5-mm-wide bin were grouped irrespective of their anteroposterior coordinates. For each bin, starting at a distance of 0.5 mm from the midline and extending 3.5 mm laterally, the number of vibrissa sites were tallied and converted to frequency by expressing data as a percentage of the total number of vibrissa movement sites. Statistical differences between groups of hemispheres were determined using ANOVA (P<0.05) (see Results). Results In seven hemispheres of the control rats, 395 penetrations of the frontal cortex were made to map vibrissal movement representation. The general features of the size, shape and location of vibrissal representation in the control rats conform to previous descriptions of the rat M1 (for review, see Wise and Donoghue 1986). In general, the vibrissa representation is an anteroposteriorly elongated strip medial to the forelimb region, extending from 0.5 mm posterior to 3.5 mm anterior to the bregma and from 1 to 3.5 mm lateral to the bregma (Fig. 2A, squared solid symbols, line enclosed). Hindlimb representation delimited the posterior boundary of the vibrissa representation. In the strip of frontal cortex situated medially to vibrissa representation, it was either impossible to evoke any movement at all or meiosis and, less commonly, eyes movement was induced under the chosen stimulation

4 183 Fig. 2 Representative surface view maps of vibrissa movement representation in Control (A) and Sham (B) hemispheres. In these M1 mapping figures frontal poles are at the bottom. 0 corresponds to the bregma, and numbers indicate rostral or caudal distance from the bregma or lateral distance from the mid-line. Symbols denote stimulation threshold points evenly spaced at 0.5-mm intervals. Movements evoked at one point are indicated by symbols: filled square line enclosed, vibrissae; empty diamonds, forelimb; crosses, hindlimb; empty triangle, neck-upper trunk; filled stars, jaw-tongue; empty circles, eye; small filled points, sites unresponsive at 60 µa; absence of symbol within or at the border of the maps: penetration not performed in correspondence of large vessel conditions. Non-responsive points formed the basis for delineating the rostral border of the vibrissa region. The movement representations around vibrissa representation were not mapped in sufficient detail to determine their sizes (Fig. 2A, empty symbols). In the sham group of animals, 427 microelectrode penetrations were made in seven hemispheres to map vibrissa movement representation in the frontal cortex contralateral to the side that underwent surgery. The mean size of the vibrissa representation in sham hemispheres did not prove significantly different from that of control hemispheres (sham hemispheres mean size: 5.5 mm 2 ±0.7, range: mm 2, versus normal hemispheres mean size: 4.9 mm 2 ±1, range: mm 2 ; P=0.29, t-test; Fig. 2A, B; Fig. 5). In the deafferented groups of animals, 399 microelectrode penetrations were made in the frontal cortices of seven hemispheres contralateral to deafferented vibrissa and 419 penetrations were made into the frontal cortices of seven hemispheres in animals that underwent bilateral vibrissa deafferentation. Vibrissa movements were evoked reliably from the hemispheres of both groups of animals after 2 weeks of vibrissa deafferentation. The type of vibrissa movement evoked by microstimulation applied to hemispheres of experimental animals appeared similar to those observed in control and sham groups of animals. Quantitative comparisons of the sizes of the vibrissa representations verified that, in spite of individual variability, as evidenced in Fig. 3 and Fig. 4, there were no significant differences in the mean size of vibrissa repre-

5 184 Fig. 3 A C Representative surface view maps of vibrissa movement representation in hemispheres contralateral to unilateral infraorbital nerve damage. Maps illustrate the individual variation in size and configuration of vibrissa movement representation after contralateral vibrissal pad deafferentation sentation across deafferented groups (P=0,95, t-test; see Fig. 5). Moreover, the size of the M1 vibrissa representation area in the hemispheres of both deafferented groups of animals appeared not significantly different from those of the control and sham groups (ANOVA: F=0.53; P=0.67; see Fig. 5). To verify whether vibrissa deafferentation affected the relative mediolateral positioning of Fig. 4 A C Representative surface view maps of vibrissa movement representation in bilateral deafferented rats. Maps illustrate the individual variation in size and configuration of vibrissa movement representation after bilateral vibrissal pad deafferentation the vibrissa representation, the mean percentages of penetrations from which vibrissa movement was evoked were plotted as a function of mediolateral position, irrespective of anteroposterior coordinate (see Materials and methods). Figure 6 demonstrates for each group the relative mediolateral distribution of penetrations from which vibrissa movement was evoked. Statistical comparisons across the four groups revealed no significant differences between any of the groups in the relative mediolateral positioning of the vibrissa representations (ANOVA: F=0.08; P=0.97).

6 Fig. 5 Sizes of vibrissa representations in hemispheres of the four groups of rats. Data represent mean values (+SD). The difference in mean size of vibrissa representation between four groups of hemispheres was not statistically significant (P=0.67, ANOVA) 185 Microstimulation of the vibrissal motor region in deafferented groups of animals led to the most striking effects of infraorbital nerve lesion: an increase in the threshold current was required to evoke vibrissa movement. The lowest threshold current vibrissa sites (<15 A) were not found in the unilateral and bilateral input-deprived hemispheres. The mean threshold required to evoke vibrissa movement in unilateral and bilateral input-deprived hemispheres was significantly higher than the mean threshold required in control and sham hemispheres (ANOVA: F=22.1; P=<0.01; see Fig. 7A). In contrast, there were no significant differences between unilateral and bilateral input-deprived hemispheres in the mean threshold required to evoke vibrissa movement (P=0.46, t-test; see Fig. 7A). Modifications in excitability in the input-deprived hemispheres were also seen in the plots indicating the distribution of threshold currents required to induce vibrissa movements in each group (Fig. 8A). Vibrissa sites with thresholds below 20 A resulted over 30% of all excitable sites in control and sham hemispheres but below 10% of all excitable sites in unilateral and bilateral input-deprived hemispheres. To analyze mediolateral threshold variability in input deprived vibrissa representations, the thresholds required to evoke vibrissa movements were examined by computing, for each animal, the mean current used to evoke these movements versus their midline, irrespective of the anteroposterior coordinates. The mediolateral plot of mean thresholds for each group of hemispheres is reported in Fig. 9. In both the control and sham hemispheres, throughout mediolateral extent of the M1 vibrissa representation, vibrissa movements were evoked at thresholds lower than those required in unilateral and bilateral deprived hemispheres (ANOVA, F=24.5; P<0.01) This shows that the vibrissa representation excitability in input-deprived hemispheres was lower than normal in all mediolateral positions. The effect of vibrissa deafferentation on current threshold appeared to be restricted to the vibrissa representation because, between hemispheres and between groups, there were no differences in the mean current threshold required to evoke other types of movements (Fig. 7B D). Fig. 6 Comparison of the mediolateral frequency distribution of penetrations eliciting vibrissa movement between four groups of hemispheres. Statistical comparisons across groups for each bin revealed no significant differences

7 186 Fig. 8 Distribution of thresholds for vibrissa (A) and forelimb (B) movement in the four groups of hemispheres. Each histogram has been normalized for the total number (N) of effective sites. The unilateral and bilateral groups of deafferented hemispheres evidenced a large decrease in the percentage of sites with threshold <20 µa and an increase in the percentage of sites with threshold current of 30 µa. These differences in threshold distribution between groups are not present in distribution of threshold for forelimb movement Fig. 7 Current threshold for evoking movements from the four groups of hemispheres. Data represent mean value (+SD). Current threshold for evoking vibrissa movement (A) from both unilateral and bilateral deafferented groups was significantly higher than for the control and sham groups (P<0.01, ANOVA). Across-group comparisons of current thresholds for evoking other types of movement (B D) were not statistically significant Discussion This study was designed to identify the contribution of vibrissal pad sensory input in maintaining the M1 representation pattern. The finding of the present experiment is that the ablation of vibrissal pad input did not result in size and/or topographic modifications in the contralateral vibrissa motor representation, although it did increase the mean current threshold required to evoke movements from deafferented vibrissa. The results of the present study evidenced a significant increase in the threshold required to evoke vibrissa movement in deafferented groups versus that required in comparison with control and sham groups. Such increases in threshold were specific because there were no acrossgroup differences in the current threshold required to evoke movements from other body parts.

8 187 Fig. 9 Mediolateral distribution of thresholds for vibrissa movement in the four groups of hemispheres. Thresholds were grouped according to their distance from the midline irrespective of the anteroposterior coordinate. Throughout the mediolateral extension of vibrissa representation, higher thresholds were required to evoke vibrissa movement in both unilateral and bilateral groups of hemispheres. In the control group of hemispheres only at the 3.5 mm of mediolateral coordinate, the mean vibrissa threshold value is higher than those observed in other sites. Nevertheless, at this mediolateral coordinate, vibrissa movement at threshold current was seen only in four sites out of seven hemispheres Chronic peripheral nerve injuries produce neural changes at different levels of the somatosensory system (for review, see Kaas 1999). As a general rule, after normal sensory input has been removed, the corresponding sensory cortex is initially silenced (or presumed silent); during the post-lesional period (weeks to months), this cortex adopts the functional properties of its immediate cortical surroundings. In one of the early studies on denervation-induced plasticity, permanent transection of the vibrissae sensory nerve (ION) was carried out as in the present experiment, with regeneration being prevented. It was found that, in animals deafferented as adult, no evidence of plasticity could be detected in the trigeminal nuclei and limited reactivation was apparent in the barrel sensory cortex, so that after 60 days, 75% of the disconnected region remained unresponsive to peripheral stimulation (Wite 1984). Comparable results were also described recently by Kis et al. (1999). In this study, responses evoked by ipsilateral vibrissal stimulation in the trigeminal nucleus reappeared only days after the ION had been transected without suturing. In the ventral postero-medial nucleus of the thalamus (VPM) the vibrissal representation areas were unresponsive to stimulation of both the vibrissae and other parts of the face until regeneration had occurred. In the barrel cortex of rodents, a partial vibrissectomy performed in adult animals induces expansion of functional representation of the spared vibrissae (Kossut et al. 1988; Suicinska and Kossut 1994). However, the functional representation of spared vibrissae did not occupy the entire denervated barrel field (Suicinska and Kossut 1994). In contrast, immediate and simultaneous sensory reorganization was observed at all levels of the somatosensory system when a reversible sensory deactivation was produced by subcutaneous injection of lidocaine (Faggin 1997). However, it is noteworthy that novel responses have also been ob- served shortly after digit denervation in the SI (Calford and Tweedale 1988; Byrne and Calford 1991). In this light, during the 2 weeks of survival after ION deafferentation in the present experiment, the main somatosensory input to the vibrissal motor cortex came from silent areas or areas which had undergone limited reorganization. Vibrissa motor representation receives extensive associative connections from S1 and S2 (Donoghue and Parham 1983; Izraeli and Porter 1995) that represent the main sources of short-latency somatosensory input to M1 (Farkas et al 1999). Moreover, direct somatosensory feedback could also reach the vibrissa motor cortex directly through thalamic nuclei: VL and PO (Donoghue et al. 1979; Herkenham 1980; Diamond et al. 1992). Studies, using magnetic brain stimulation in humans, have demonstrated that the manipulation of afferent input can produce alterations in motor cortical excitability (Ridding et al. 1999) and motor cortical plasticity (Brasil- Neto et al. 1993; Hamdy et al. 1998; Zeimann et al. 1998a, 1998b). The increase in threshold required to evoke vibrissa movement, subsequent to the subtraction of vibrissal sensory modulation upon cerebral cortical motor output, most likely reflects persistent plastic changes at the cortical level and/or within the terminal structures the vibrissa motor cortex projects into (Porter and White 1983; Miyashita et al. 1994). Thus the higher than normal threshold for all mediolateral vibrissa representation sites may reflect a uniform effect on connectivity within the vibrissa representation. This effect on cortical circuits is persistent and cannot be compensated by other input to the vibrissa representation. The results of the present study do not provide any direct evidence as to the mechanisms implicated in mediating changes in the cortical circuits. Nevertheless, the effects upon motor cortical circuits of such abnormal input could be a change in the synaptic efficacy of pre-existing

9 188 connections (Keller 1993; Donoghue et al. 1996) or changes in synapse number (Kleim et al. 1996). It has been suggested that whisking, like other rhythmic motor behaviors, may be affected by a network of neurons that function as a central pattern generator (Sharp and Evans 1982; Isokawa-Akerson and Komisaruk 1987). The motor cortex plays a role in initiating whisking and in modulating the level and temporal activity of vibrissa muscles (Carvel et al. 1996). The role of the vibrissa feedback loop to motor cortex in whisking behaviour is still unclear and the functional significance of elevated electrical stimulation threshold after vibrissa deafferentation is not clarified by the results of the present experiment. Since the motor cortex appeared involved in the high degree of whisking control (Carvel et al. 1996), the increase in electrical stimulation threshold after deafferentation could be an electrophysiological correlate of the impairment of sensorimotor coordination of whisking observed in deafferented rats. The influence of ipsilateral vibrissa input upon ipsilateral sensorimotor cortex is well documented (Armstrong- James and George 1988; Calford and Tweedale 1990; Clarey et al. 1996). Comparison of the effects of unilateral and bilateral deafferentation evidenced that the subtraction of ipsilateral input to vibrissa motor representation apparently does not change the excitability of vibrissa motor representation. It may be that subtle effects on thresholds go undetected under the stimulation conditions used in this study. In contrast to effects on adult map motor cortex reorganization induced by facial nerve lesion (Sanes et al 1990; Toldi et al 1996; Huntley 1997a) or facial nerve reinnervation (Franchi 2000), there was no evidence that vibrissa deafferentation led to modifications of the M1 pattern representation. The results of the present experiment are in agreement with previous observations that the motor cortices of adult animals are more resistant to the manipulation of vibrissa input than at birth (Carvel and Simons 1996; Huntley 1997b). The fact that the effects of whisker deafferentation does not affect size and topographical relationship of vibrissa movement representation provides support for the conclusion that the form of long-term plasticity examined in the present study is different from that induced by facial motor nerve lesions (Sanes et al. 1990; Toldi et al. 1996) and facial nerve reinnervation (Franchi 2000). The present results may suggest that the form of long-term M1 plasticity is specifically linked to the type of peripheral damage and that persistent reshaping of M1 reorganization is strictly triggered by motor output damage. Therefore it may be that the mechanisms underlying the changes induced by peripheral sensory nerve damage are distinct from those underlying changes induced by motor nerve damage. The plastic adaptation of cortical circuits after vibrissa input deprivation could reflect long-term changes in the functional efficacy of circuits limited to within the vibrissa representation (Keller 1993; Hess and Donoghue 1994). On the other hand, plastic adaptation after facial motor disconnection could reflect long-term changes in the longer distance circuits that have been proposed for border shifts after peripheral nerve lesion (Jacobs and Donoghue 1991; Donoghue 1995; Huntley 1997a). The current influential view is that the stability of representation borders depends on GABA-ergic function of cortical inhibitory interneurons. This view was suggested by the finding that pharmacological blockade of GABA receptors by local application of bicuculline into rat forelimb motor representation led to an expansion of forelimb movement in the vibrissa representation (Jacobs and Donoghue 1991). This effect was interpreted as secondary to the unmasking of pre-existing excitatory connections from vibrissae to forelimb motor cortex, normally suppressed by local inhibitory circuits. Thus, the present results may suggest that vibrissa deafferentation did not induce a persistent down-regulation of these local inhibitory interneurons together with the unmasking of pre-existing excitatory connections from vibrissa to forelimb motor cortex. Nevertheless, the present experiment does not rule out the possibility that some other set of non dominant inputs from the face or neck may be relayed to the vibrissa motor region, contributing to keeping vibrissal size representation normal in deafferented animals (Schroeder et al. 1995; Moore and Nelson 1998). In conclusion, the model of deafferentation-induced plasticity used in the present study suggests that the spatial extent of the long-term reorganization in deafferented M1 appears limited to vibrissa representation, without involvement of neighboring representations. However, future studies will need to establish whether this conclusion holds only for the vibrissal motor system organization or whether it can be extended as a general feature of motor system representations in adult mammals. Acknowledgements I thank Dr. C. Lucchetti for manuscript revision and Ing. E. Lodi for assistance during experiments. I am also grateful to Mr. V. Muzzioli for his assistance with preparation of the figures. This study was funded by grants from Ministero dell Università e della Ricerca Scientifica e Tecnologica (MURST). References Armstrong-James M, George MJ (1988) Bilateral receptive fields of cells in rat Sm1 cortex. Exp Brain Res 70: Archibald SJ, Kraarup C, Shefner J, Li S, Madison RD (1991) A collagen-based nerve guide conduit for peripheral nerve repair: an electrophysiological study of nerve regeneration in rodents and nonhuman primates. J Comp Neurol 306: Brasil-Neto JP, Valls-Sole J, Pascual-Leone A, Cammarota A, Amassian VE, Cracco R, Maccabee P, Cracco J, Hallett M, Cohen LG (1993) Rapid modulation of human cortical motor outputs following ischaemic nerve block. Brain 116: Byrne JA, Calford MB (1991) Short-term expansion of receptive fields in rat primary somatosensory cortex after hindpaw digit denervation. Brain Res 565: Calford MB, Tweedale R (1988) Immediate and chronic changes in responses of somatosensory cortex in adult flying-fox after digit amputation. Nature 332: Calford MB, Tweedale R (1990) Interhemispheric transfer of plasticity in the rat cerebral cortex. Science 249: Carvel GE, Simonds DJ (1996) Abnormal tactile experience early in life disrupt active touch. J Neurosci 16:

10 189 Carvel GE, Miller SA, Simonds DJ (1996) The relationship of vibrissa motor cortex unit activity to whisking in the awake rat. Somat Motor Res 13: Clarey JC, Tweedale R, Calford MB (1996) Interhemispheric modulation of somatosensory receptive fields: evidence for plasticity in primary somatosensory cortex. Cereb Cortex 6: Diamond ME Armstrong-James MA, Budway MJ, Ebner FF (1992) Somatic sensory responses in the rostral sector of the posterior group (POm) and in the ventral posterior medial nucleus (VPM) of the rat thalamus: dependence on the barrel field cortex. J Comp Neurol 319:66 84 Donoghue JP (1995) Plasticity of adult sensorimotor representation. Curr Opin Neurobiol 5: Donoghue JP, Parham C (1983) The afferent connections of the lateral agranular field of the rat motor cortex. J Comp Neurol 217: Donoghue JP, Wise SP (1982) The motor cortex of the rat: cytoarchitecture and microstimulation mapping. J Comp Neurol 212:76 88 Donoghue JP, Kerman KL, Ebner FF (1979) Evidence for two organizational plans within the somatic sensory-motor cortex of the rat. J Comp Neurol 183: Donoghue JP, Suner S, Sanes JN (1990) Dynamic organization of primary motor cortex output to target muscles in adult rats. II. Rapid reorganization following motor nerve lesion. Exp Brain Res 79: Donoghue JP, Hess G, Sanes JN (1996) substrates and mechanisms for learning in motor cortex. In: Bloedel J, Ebner T, Wise SP (eds)acquisition of motor behavior. MIT, Cambridge, Mass., pp Faggin BM, Nguyen KT, Nicolelis MAL (1997) Immediate and simultaneous sensory reorganization at cortical and subcortical levels of the somatosensory system. Proc Natl Acad Sci USA 94: Farkas T, Kis ZS, Toldi J, Wolff JR (1999) Activation of the primary motor cortex by somatosensory stimulation in adult rats is mediated mainly by associational connections from the somatosensory cortex. Neuroscience 90: Franchi G (2000) Reorganization of vibrissal motor representation following severing and repair of the facial nerve in adult rats. Exp Brain Res 131:33 43 Hamdy S, Rothwel JC, Aziz Q, Singh KD, Thompson DG (1998) Long-term reorganization of human motor cortex driven by short-term sensory stimulation. Nature Neurosci 1:64 68 Herkeman M (1980) Laminar organization of thalamic projections to the rat motor cortex. Science 207: Hess G, Donoghue JP (1994) Long-term potentiation of horizontal connections provides a mechanism to reorganize cortical motor maps. J Neurophysiol 71: Huntley GW (1997a) Correlation between pattern of horizontal connectivity and the extent of short-term representational plasticity in rat motor cortex. Cereb Cortex 7: Huntley GW (1997b) Differential effects of abnormal tactile experience on shaping representation patterns in developing and adult motor cortex. J Neurosci 17: Isokawa-Akerson M, komisaruk BR (1987) Difference in projections to the lateral and medial facial nucleus: anatomically separate pathway for rhythmical movement in rats. Exp Brain Res 65: Izraeli R, Porter LL (1995) Vibrissal motor cortex in the rat: connections with the barrel field. Exp Brain Res 104:41 54 Jacobs KM, Donoghue JP (1991) Reshaping the cortical motor map by unmasking latent intracortical connections. Science 251: Kaas JH (1999) The reorganization of sensory and motor maps after injury in adult mammals. In: Gazzaniga MS (ed) The new cognitive neurosciences. MIT, Cambridge, Mass., pp Kaneko T, Caria MA, Asanuma H (1994) Information processing within the motor cortex. I. Responses of morphologically identified motor cortical cells to stimulation of the somatosensory cortex. J Comp Neurol 345: Keller A (1993) Intrinsic synaptic organization of the motor cortex. Cereb Cortex 3: Keller A, Weintraub D, Miyashita E (1996) Tactile experience determines the organization of movement representations in rat motor cortex. NeuroReport 7: Kis Z, Farkas T, Rabl K, Kis E, Korodi K, Simon L, Marusin I, Rojik I, Toldi J (1999) Comparative study of the neuronal plasticity along the neuraxis of the vibrissal sensory motor system of the adult rat following unilateral infraorbital nerve damage and subsequent regeneration. Exp Brain Res 126: Kleim JA, Lussing E, Schwarz Er, Comery TA, Greenough WT (1996) Synaptogenesis and FOS expression in the motor cortex of the adult rat motor skill learning. J Neurosci 16: Kossut M, Hand PJ, Greenberg J, Hand CL (1988) Single vibrissa cortical column in SI cortex of rat and its alterations in neonatal and adult vibrissa-deafferented animals: a quantitative 2-DG study. J Neurophysiol 60: Miyashita E, Keller A, Asanuma H (1994) Input-output organization of the rat vibrissal motor cortex. Exp Brain Res 99: Moore CI, Nelson SB (1998) Spatio-temporal subtreshold receptive fields in the vibrissa representation of the rat primary somatosensory cortex. J Neurophysiol 80: Neafsey EJ, Bold EL, Haas G, Hurley-Gius KM, Quirk G, Sievert CF, Terreberry RR (1986) The organization of the rat motor cortex: microstimulation mapping study. Brain Res 396:77 96 Porter LL (1996) Somatosensory input onto pyramidal tract neurons in rodent motor cortex. NeuroReport 7: Porter LL, White EL (1983) Afferent and efferent pathways of the vibrissal region of primary motor cortex in the mouse. J Comp Neurol 214: Ridding MC, Rothwell JC (1999) Afferent input and cortical organization: a study with magnetic stimulation. Exp Brain Res 126: Sanes JN, Donoghue JP (1997) Static and dynamic organization of motor cortex. Adv Neurol 73: Sanes JN, Suner S, Lando JF, Donoghue JP (1988) Rapid reorganization of adult rat motor cortex somatic representation patterned after motor nerve injury. Proc Natl Acad Sci USA 85: Sanes JN, Suner S, Donoghue JP (1990) Dynamic organization of primary motor cortex output to target muscles in adult rats. I. Long-term pattern of reorganization following motor or mixed peripheral nerve lesion. Exp Brain Res 79: Sanes JN, Wang J, Donoghue JP (1992) Immediate and delayed changes of rat motor cortical output representation with new forelimb configurations. Cereb Cortex 2: Schroeder CE, Seto S, Arezzo JC, Garraghty PE (1995) Electrophysiological evidence for overlapping dominant and latent inputs to somatosensory cortex in squirrel monkeys. J Neurophysiol 74: Sharp FR, Evans K (1982) Regional (14 C)2-deoxiglucose uptake during vibrissae movements evoked by rat motor cortex stimulation. J Comp Neurol 208: Sievert CF, Neafsey EJ (1986) A chronic unit study of the sensory properties of neurons in the forelimb areas of the rat sensorimotor cortex. Brain Res 381:15 23 Suicinska E, Kossut M (1994) Short-term changes of cortical body maps following partial vibrissectomy in adult mice. Acta Neurobiol Exp 54: Toldi J, Laskawi R, Landgrebe M, Wolff JR (1996) Biphasic reorganization of somatotopy in the primary motor cortex follows facial nerve lesion in adult rats. Neurosci Lett 203: Wise SP, Donoghue JP (1986) Motor cortex of rodents. In: Jones EG, Peters A (eds) Cerebral cortex, vol 5. Sensory motor areas and aspects of cortical connectivity. Plenum, New York, pp Wite PEM (1984) Rearrangement of neural responses in the trigeminal system of the rat following peripheral nerve section. J Physiol 352: Zeimann U, Corwell B, Cohen LG (1998a) Modulation of plasticity in human motor cortex after forearm ischaemic nerve block. J Neurosci 18: Zeimann U, Hallett M, Cohen LG (1998b) Mechanisms of deafferentation-induced plasticity in human motor cortex. J Neurosci 18: