* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project
Download PII: S0006-8993(97) - UCSD Cognitive Science
Document related concepts
Brain Research 769 Ž1997. 256–262 Research report Denervation-induced sprouting of intact peripheral afferents into the cuneate nucleus of adult rats D.R. Sengelaub a,b , N. Muja a , A.C. Mills c , W.A. Myers a , J.D. Churchill a,b , P.E. Garraghty a,b,) a Department of Psychology, Indiana UniÕersity, Bloomington, IN 47405, USA Program in Neural Science, Indiana UniÕersity, Bloomington, IN 47405, USA Department of Biology, Middle Tennessee State UniÕersity, Murfreesboro, TN 37132, USA b c Accepted 28 May 1997 Abstract In adult monkeys with dorsal rhizotomies extending from the second cervical ŽC 2 . to the fifth thoracic ŽT5 . vertebrae, cortex deprived of its normal inputs regained responsiveness to inputs conveyed by intact peripheral afferents from the face wT.P. Pons, P.E. Garraghty, A.K. Ommaya, J.H. Kaas, E. Taub, M. Mishkin, Massive reorganization of the primary somatosensory cortex after peripheral sensory deafferentation, Science 252 Ž1991. 1857–1860x. It has been suggested that the extent of this massive topographic reorganization may be due to the establishment of novel connections between intact afferents and neurons denervated after dorsal rhizotomy wP.E. Garraghty, D.P. Hanes, S.L. Florence, J.H. Kaas, Pattern of peripheral deafferentation predicts reorganizational limits in adult primate somatosensory cortex, Somatosens. Motor Res. 11 Ž1994. 109–117x. Using adult rats with comparably extensive dorsal rhizotomies, we employed anatomical tracing techniques to address this possibility. Subcutaneous hindpaw injections of horseradish peroxidase conjugated to either wheat germ agglutinin or cholera toxin subunit B revealed aberrant expansions of gracile projections into the cuneate and, in one case, external cuneate nucleus within three months of the deafferentation. It seems plausible that such modest sprouting of ascending projections at the level of the brainstem may form functional connections which, through divergence, ultimately drive a larger population of neurons in cortex. This new growth may well account for both the substantial cortical reorganization observed in the ‘Silver Spring monkeys’ wT.P. Pons, P.E. Garraghty, A.K. Ommaya, J.H. Kaas, E. Taub, M. Mishkin, Massive reorganization of the primary somatosensory cortex after peripheral sensory deafferentation, Science 252 Ž1991. 1857–1860x and the ‘referred sensation’ phenomena Žsee J.P. Donoghue, Plasticity of adult sensorimotor representations, Curr. Opin. Neurobiol., 5 Ž1995. 749–754 for review. reported to follow proximal limb amputations in humans. q 1997 Elsevier Science B.V. Keywords: Dorsal rhizotomy; Brainstem; Gracile nucleus; Sprouting; Deafferentation 1. Introduction In 1991, Pons et al. w17x reported a reorganization of somatosensory cortex in adult macaques with complete deafferentations of the arm, hand, and upper trunk that was as much as an order of magnitude greater than cortical reorganizations found after peripheral injuries w3,12–14x. In the so-called ‘Silver Spring monkeys’, a circumscribed portion of the representation of the face was found to have expanded in excess of 10 mm medially, apparently across the entire deprived region of somatosensory cortex. ) Corresponding author. Department of Psychology, Indiana University, Bloomington, IN 47405, USA. Fax: q1 Ž812. 855-4520; E-mail: [email protected] 0006-8993r97r$17.00 q 1997 Elsevier Science B.V. All rights reserved. PII S 0 0 0 6 - 8 9 9 3 Ž 9 7 . 0 0 7 0 8 - 7 Unlike topographic changes seen after peripheral nerve damage w2,12,13x, the extent of the topographic reorganization found in the Silver Spring Monkey was sufficiently large that it could not be readily accounted for based on the maximal sizes of thalamocortical axonal arbors w7,19x. The apparent insufficiency of existing anatomical connections to account for the extensive rhizotomy-induced reorganization suggested that novel connections may well have been established. Rausell et al. w18x suggested that relay neurons in the ventroposterior medial nucleus ŽVPM. might have sprouted novel connections onto denervated cells in the ventroposterior lateral nucleus ŽVPL., and that the trigeminal receptive fields represented in VPM were then conveyed to ‘VPL cortex’ over this new pathway. We found it more plausible to imagine that new growth had indeed occurred, but that it involved the sprouting of D.R. Sengelaub et al.r Brain Research 769 (1997) 256–262 primary peripheral afferents at the level of the brainstem since vacated synaptic sites no doubt existed in the denervated cuneate nucleus Žsee Fig. 6B,C.. It seemed reasonable to hypothesize that modest sprouting at the level of the brainstem could, through divergence, innervate a larger population of neurons in the thalamus which could consequently drive an even larger population of neurons in the cortex. To evaluate the hypothesis that intact afferents may form novel connections with denervated targets in the brainstem, we have performed unilateral dorsal rhizotomies in adult rats as extensive as those in the monkeys studied by Pons et al. w17x. Our results provide evidence for denervation-induced, sprouting of intact peripheral afferents into the denervated cuneate nucleus of adult rats. A preliminary report of some of these data has appeared elsewhere w6x. 2. Materials and methods Thirteen adult, Sprague–Dawley rats weighing between 200 and 360 g underwent unilateral rhizotomy of the dorsal roots projecting to the left cuneate nucleus. Animals were deeply anesthetized with an intramuscular Ži.m.. injection of ketamine HCl ŽKetaset, 80 mgrkg. and xylazine ŽRompun, 10 mgrkg.. Supplemental doses were administered as needed in order to maintain a surgical level of anesthesia. Their backs were shaved and prepared for surgery with alternate scrubbings of Betadine and alcohol. The animals were placed on a heating pad to maintain normal body temperature and ophthalmic ointment was applied to prevent corneal desiccation. A skin incision was made along the dorsal midline from the base of the skull to the mid-spinal level. The layers of muscle overlying the spinal column were incised and retracted. Under microscopic view, fine rongeurs were used to perform partial laminectomies from the second cervical to the fourth or fifth thoracic vertebrae thereby exposing the spinal cord. The dura encasing the spinal cord was incised and reflected to expose the dorsal roots for avulsion. After dorsal root transection, Gelfilm and a pledget of Gelfoam ŽUpjohn. were placed over the exposed cord for protection and support. The overlying musculature was then sutured in layers and the skin wound was sutured shut. Postoperatively, the rats received i.m. injections of dexamethasone Ž0.8 mgrkg., Dopram Ž8 mgrkg., penicillin ŽAmbi-Pen, 120 000 unitsrkg., and lactated ringers solution with 5% dextrose Ž5% body weightrh of surgery, s.c... A bolus of penicillin was administered daily for 3 days. Recovery from these procedures was generally uneventful, though in 2 of the 13 rats, some autotomy was observed. After survival durations ranging from 58 to 304 days, the rats were anesthetized with ether. Horseradish peroxidase conjugated with either: Ž1. cholera toxin subunit B Ž1% B-HRP, Sigma.; Ž2. wheat germ agglutinin Ž2% 257 WGA-HRP, type VI, Sigma.; or Ž3. a combination of both was dissolved in distilled water and injected subcutaneously into several bilaterally matched areas of the animals hindpaws Ž n s 5; and face, n s 2 of these 5., or into the hindpaw on the deafferented side Ž n s 8. using a Hamilton microsyringe. One additional normal control rat received unilateral hindpaw injections of tracer. There are, thus, 13 cases of hindlimb injections ipsilateral to a denervated cuneate nucleus, and 6 cases with hindlimb injections ipsilateral to an intact cuneate nucleus. Three days were allowed for the anatomical tracerŽs. to be transported anterogradely to the central nervous system. At that time, the rats were administered a lethal overdose of urethane Ž0.5 mgrml, i.p.. and perfused transcardially through the ascending aorta at normal blood pressure with physiological saline followed by a cold mixture of 1% paraformaldehyde and 1.25% glutaraldehyde. The brainstems were then carefully extracted and postfixed for 5 h in the same fixate and transferred to 0.1 M phosphate buffer ŽpH 7.4. containing 10% sucrose overnight at 48C. The tissue was embedded in gelatin and sectioned coronally or, in 4 of the experimental cases, horizontally, at 40 mm on a freezing microtome. Sections were cut into cold phosphate buffer ŽpH 7.4. and processed histochemically, according to Mesulam’s w15x protocol, with tetramethylbenzidine ŽTMB, Sigma. as the chromagen. In some cases, alternate sections were counterstained with 1.0% Neutral red. The sections were mounted on gelatinized slides and allowed to dry overnight. The tissue was dehydrated in increasing concentrations of alcohol, cleared in a series of xylenes, and coverslipped with Permount. Camera lucida reconstructions of the anatomical location of the labeling were done under darkfield at 10 = magnification. Terminal label was reconstructed in brightfield at 100 = magnification with an oil immersion objective. Nuclear boundaries were defined using the atlas of Paxinos and Watson w16x. Additional quantitative measurements were made. First, we counted the number of instances of terminal label in the cuneate nucleus after hindlimb tracer injection. Clearly separated patches of terminal label within and across sections were counted as single instances of aberrant terminations. Second, the mean area of the aberrant terminal label was calculated by taking the average of the areas of minimal convex polygons drawn to include all of the aberrant terminal label in each section of each case. 3. Results Phenotypically, the success of the surgery was evidenced by a complete lack of use of the deafferented forelimb in locomotion Že.g., see w24x.. Interestingly, the deafferented forelimb was used in stereotypical bilateral grooming and feeding behaviors suggesting that forelimb use in these behaviors requires no afferent feedback from 258 D.R. Sengelaub et al.r Brain Research 769 (1997) 256–262 the forelimb for their initiation and maintenance Žthough it is possible that stimulation of the head and face by the deafferented forelimb provides inputs that could shape the stereotypical forelimb behavior.. Tissue reacted for HRP using TMB revealed aberrant sprouting of intact gracile projections into denervated cuneate nuclei in 12 of the 13 subjects within 2 months of the deafferentation. Fig. 1A,B shows low- and high-power micrographs of coronally sectioned tissue taken from one case. A serial reconstruction of the terminal label Žacross 3 sections. is illustrated in Fig. 2. As can be seen, there is an unambiguous expansion of label into the cuneate nucleus extending from the vicinity of the gracile nucleus. Fig. 1C,D and Fig. 3 illustrate comparable data from another subject. In this case, the brainstem was sectioned in the horizontal plane. Again, the photomicrographs ŽFig. 1C,D. demonstrate that HRP-labeled fibers extended into the denervated cuneate nucleus. The camera lucida reconstruction of this label is illustrated in Fig. 3. Comparable aberrant label was found in the denervated cuneate nucleus in 12 of the 13 subjects whether the anatomical tracer was injected into both face and hindlimb Ž n s 2. or hindlimb Fig. 1. Darkfield photomicrographs of HRP terminal labeling in the brainstems in two rats that had undergone dorsal rhizotomies extending from C 2 to T5 . In both cases, the anatomical tracer was injected into the hindlimb ipsilateral to the deafferentation. A: coronal section showing label in the gracile nucleus Žopen arrow. and the anomalous labeling in the cuneate nucleus Žsolid arrow.. Medial is to the right, dorsal to the top. B: higher magnification photomicrograph of anomalous labeling shown in A. C: horizontal section showing labeling in the gracile nucleus Žopen arrow. and anomalous labeling in the cuneate nucleus Žsolid arrow.. Medial is to the right, rostral to the top. D: higher magnification photomicrograph of the anomalous labeling shown in C. Scale bars Žin A,C. s 250 mm; Žin B,D. s 100 mm. D.R. Sengelaub et al.r Brain Research 769 (1997) 256–262 259 Fig. 2. Camera lucida reconstruction of terminal label illustrated in Fig. 1A,B. This label was serially reconstructed across three 50-mm sections. The labeling in the cuneate nucleus clearly appears to extend from the gracile nucleus. AP, area postrema; A2, A2 noradrenergic cells. alone Ž n s 11.. It is, therefore, probable, that the cuneate label in all cases derived from the hindlimb injection sites. While not illustrated, we saw no such innervation of the cuneate nucleus by gracile fibers in any of the 5 rats in which matched injections of anatomical tracer were made contralateral to the rhizotomy, nor was any cuneate labeling detected in the one normal control rat. Thus, the histological label found in the cuneate nucleus ipsilateral to the rhizotomy in 12 of the 13 rhizotomized subjects is apparently not characteristic of typical hindlimb projections to the brainstem, but, rather reflects opportunistic Fig. 3. Camera lucida reconstruction of the terminal label illustrated in Fig. 1C, D. Extensive labeling is again evident in the cuneate nucleus. Fig. 4. Histograms representing the mean Ž"S.E.M.. areal extent of the aberrant terminal label in the cuneate nucleus ipsilateral to the dorsal rhizotomies after the injection of anatomical tracer into the ipsilateral hindpaw. The left-hand bar is for coronally sectioned material; the right-hand bar is for horizontally sectioned material. Control data are not represented as there were no instances of such label in any of those cases. 260 D.R. Sengelaub et al.r Brain Research 769 (1997) 256–262 Fig. 5. Histograms representing the mean Ž"S.E.M.. number of discrete patches of terminal label in the cuneate nucleus ipsilateral to the dorsal rhizotomies after the injection of anatomical tracer into the ipsilateral hindpaw. The left-hand bar is for coronally sectioned material; the right-hand bar is for horizontally sectioned material. Control data are not represented as there were no instances of such label in any of those cases. expansion of intact afferents into the denervated nucleus Ž x 2 s 13.45, P - 0.005.. Fig. 4 presents the mean area of HRP label in the cuneate nucleus after hindlimb tracer injection. The data presented are the means from the coronally and horizontally sectioned brainstems of the rhizotomized animals. The 6 control cuneate nuclei are not represented here as we detected no aberrant terminal label in any of these animals, leaving them with a mean and standard error of 0 Žall control cases were sectioned coronally.. The difference between the coronally and horizontally sectioned cases is an artifact of the plane of section. That is, the cuneate nucleus extends over a larger area in the horizontal than in the coronal plane. Fig. 5 presents the average number of clearly separated sites of aberrant terminal label in the coronally and horizontally sectioned brainstems, and, as is evident, there was no difference. ological recording in somatosensory cortex suggests that the new connections are functional to a limited extent, as evidenced by an apparent expansion of hindlimb responsiveness into ‘cuneate’ cortex w6x. A comparable outcome has been reported for rats that had undergone neonatal forelimb amputation w10x. In those animals, the expansion of gracile fibers into the cuneate nucleus was extensive, but there was minimal invasion of the forelimb region in cortex by hindlimb inputs. The major difference between forelimb amputation and C 2 –T5 rhizotomies is that many normal inputs to the cuneate nucleus remain after amputation. Lane et al. w10x suggested that the aberrant inputs to the cuneate, which could drive activity in that nucleus, were functionally suppressed at the level of the thalamus or cortex. Such suppression might be less likely following extensive rhizotomies where the aberrant inputs are not intermingled with a larger set of ‘normal’ inputs. The incomplete invasion of the forelimbrupper trunk cortex in the rhizotomized rats relative to the apparently complete occupation of deprived cortex by inputs from the face in the Silver Spring Monkeys might be due to the substantial differences in postsurgical survival durations, or, alternatively, simply to species differences. It should be noted that small projections from gracile fasciculus fibers into the cuneate nucleus have been reported in normal rats w8,9,11,20,21x, though not consistently. For example, LaMotte et al. w9x reported a termination from the hindlimb in the cuneate nucleus in only one of 16 rats following the labelling of either the saphenous or sciatic nerve. Thus, it is possible that the aberrant label in the cuneate nucleus in the present rats reflects an expansion in a previously existing projection. In any event, label was present in 12 of the 13 experimental animals in the cuneate ipsilateral to the rhizotomies, but was not observed in the 5 rats in which matched injections were placed in the hindlimb contralateral to the deafferentation or in the one control rat. 4. Discussion 4.1. Is sprouting confined to intact peripheral afferents? In the present experiments, we have evaluated the hypothesis that aberrant growth of intact peripheral sensory afferents might follow extensive deafferentation. This hypothesis was derived from observations in adult macaque monkeys that had survived for a number of years after dorsal rhizotomies extending from C 2 to T5 w17x. In those animals, with denervated cuneate nuclei, electrophysiological mapping in primary somatosensory cortex revealed that the deprived zone of cortex had come to represent skin surfaces innervated by the trigeminal nerve w17x. The most parsimonious ‘explanation’ seemed to be that intact trigeminal fibers had formed sprouts into the denervated cuneate nucleus, and that these sprouted trigeminal afferents then used the ‘cuneate circuitry’ to relay their receptive fields to the deprived cortex w2x. The present results demonstrate clearly that new growth is possible within the adult mammalian brainstem, and preliminary electrophysi- This aberrant growth at the level of the brainstem in rhizotomized rats in no way precludes the existence of sprouting at the level of the thalamus, a possibility suggested by Rausell et al. w18x. They reported that the cuneate nuclei ipsilateral to the dorsal rhizotomies were severely shrunken, and hypothesized that this was due, at least in part, to cell loss. This appearance of a cuneate cell loss led them to suggest that VPM relay neurons whose axons course to the cortex through the VPL might have formed sprouts onto VPL neurons denervated by the loss of cuneate cells. These novel VPM connections with VPL neurons could then serve to relay trigeminal receptive fields to the ‘cuneate region’ of somatosensory cortex, providing the anatomical underpinnings for the electrophysiological data of Pons et al. w17x. We see no apparent shrinkage of the cuneate nuclei in D.R. Sengelaub et al.r Brain Research 769 (1997) 256–262 rats with equivalently extensive rhizotomies, but hasten to add that the postsurgical survival times in these rats were substantially shorter than those of the Silver Spring Monkeys Ž10–12 years.. It is certainly possible that comparable reductions in cuneate nucleus volumes would also follow more protracted survivals in the rats, but in the absence of neuronal counts the issue of cell loss must remain unresolved. Obviously, if one assumes no species differences, aberrant sprouting could occur at both brainstem and thalamic levels, but we would argue for the more parsimonious account of sprouting only at the level of the brainstem as a small expansion of intact afferents at that level could, through diverging projections, come to activate a large region of the cortex. In either case Žsprouting in brainstem and thalamus or brainstem alone., the present results demonstrate clearly that intact peripheral afferents can form novel connections central to the spinal cord in adult rats after extensive dorsal rhizotomies. 4.2. How do the present obserÕations relate to preÕiously reported data from animals with peripheral injury? Numerous experiments investigating the consequences of peripheral injuries have been performed in adults representing a variety of species Žsee w1,4x for reviews.. While there has existed Žfor some. a tacit belief that common mechanisms are operating after sensory loss arising from differing manipulations, the present results together with previous observations Že.g., w2,3,5,12–14x. suggest that different mechanisms are operating after peripheral nerve injuries and rhizotomy. So, for example, the transection of a peripheral nerve deafferents the autonomous innervation zone of that nerve, but leaves the brainstem nucleus receiving inputs conveyed by that nerve at least largely intact ŽFig. 6A.. This fact has been demonstrated in experiments employing electrical stimulation of the median, ulnar, and radial nerves in monkeys w22,23x. In these experiments a 261 ‘cryptic’ radial nerve somatosensory evoked potential to ‘median nerve cortex’ grows progressively following median nerve transection, mirroring the largely dorsum skin invasion of median nerve cortex that is reported in microelectrode mapping experiments w5,12,13x. Even after the radial nerve SEP has gained in strength Žpresumably reflecting the reorganization., electrical stimulation of the proximal median nerve stump still evokes an SEP that is comparable to that found prior to the nerve injury w23x. Thus, the central projections of the median nerve are minimally affected by transection at the mid-forearm level. Alternatively, when sensory loss is accomplished by transections proximal to the dorsal root ganglia Ži.e., rhizotomy., brainstem synaptic sites are necessarily vacated. With dorsal rhizotomies as extensive as those employed in the present experiments and in the ‘Silver Spring monkeys’, relay neurons in the cuneate nucleus are permanently disconnected from their ascending inputs ŽFig. 6B., permitting an expansion of intact afferents into the denervated cuneate nucleus ŽFig. 6C.. Moreover, in contrast to the present results, there is no evidence of intact afferent expansion in the cuneate nucleus of adult monkeys after combined median and ulnar nerve transection and subsequent reorganization ŽS.L. Florence, P.E. Garraghty, unpublished observations.. Thus, these two deafferentation paradigms differ fundamentally with regard to their central consequences, and, apparently, their central effects. In the present experiments Žand, possibly, in the ‘Silver Spring monkeys’., intact afferents form aberrant sprouts into the denervated cuneate nucleus ŽFig. 6C., presumably because molecular signals associated with denervation induce collateral growth from the neighboring population of centrally projecting fibers. While it has been previously suggested that the topographical ‘reorganizations’ that follow peripheral nerve injury arise from changes in previously existing anatomical circuitry w4x, the present results suggest that previously existing anatomy may account for little, if any, Fig. 6. Schematic illustrations of the central consequences of peripheral nerve transection ŽA. and dorsal rhizotomies extending from C 2 to T5 ŽB,C.. A: peripheral nerve transection. Dashed lines distal to the cut signify degeneration. The central projections of the affected dorsal root ganglion cells are largely unaffected. B: dorsal rhizotomy. Dashed lines signify degeneration. The central processes of dorsal root ganglion cells are affected. C: synaptic sites in the cuneate nucleus are vacated by the dorsal rhizotomies. Intact afferents from the hindlimb can then form sprouts into the denervated cuneate nucleus. 262 D.R. Sengelaub et al.r Brain Research 769 (1997) 256–262 of the central topographic changes that follow more extensive, and more proximal deafferentations. Acknowledgements This project was supported in whole or in part by B.R.S.G. Grant RR7031-27 from the Biomedical Research Support Program, Division of Research Resources, National Institutes of Health. References w1x J.P. Donoghue, Plasticity of adult sensorimotor representations, Curr. Opin. Neurobiol. 5 Ž1995. 749–754. w2x P.E. Garraghty, D.P. Hanes, S.L. Florence, J.H. Kaas, Pattern of peripheral deafferentation predicts reorganizational limits in adult primate somatosensory cortex, Somatosens. Motor Res. 11 Ž1994. 109–117. w3x P.E. Garraghty, J.H. Kaas, Large-scale functional reorganization in adult monkey cortex after peripheral nerve injury, Proc. Natl. Acad. Sci. USA 88 Ž1991. 6976–6980. w4x P.E. Garraghty, J.H. Kaas, S.L. Florence, Plasticity of sensory and motor maps in adult and developing mammals, in: V.A. Casagrande, P.G. Shinkman ŽEds.., Advances in Neural and Behavioral Development, Vol. 4, Ablex, Norwood, NJ, 1994, pp. 1–36. w5x P.E. Garraghty, N. Muja, NMDA receptors and plasticity in adult primate somatosensory cortex, J. Comp. Neurol. 367 Ž1996. 319– 326. w6x P.E. Garraghty, D.R. Sengelaub, A.C. Mills, N. Muja, Sprouting of intact afferents into the denervated cuneate nucleus of adult rats after dorsal rhizotomy, Soc. Neurosci. Abstr. 20 Ž1994. 1681. w7x P.E. Garraghty, M. Sur, Morphology of single intracellularly stained axons terminating in area 3b of macaque monkeys, J. Comp. Neurol. 294 Ž1990. 583–593. w8x G. Grant, J. Arvidsson, B. Robertson, J. Ygge, Transganglionic transport of horseradish peroxidase in primary sensory neurons, Neurosci. Lett. 12 Ž1979. 23–28. w9x C.C. LaMotte, S.E. Kapadia, C.M. Shapiro, Central projections of the sciatic, saphenous, median, and ulnar nerves of the rat demonstrated by transganglionic transport of choleragenoid-HRP ŽB-HRP. and wheat germ agglutinin-HRP ŽWGA-HRP., J. Comp. Neurol. 311 Ž1991. 546–562. w10x R.D. Lane, C.A. Bennett-Clark, N.L. Chiaia, H.P. Killackey, R.W. Rhoades, Lesion-induced reorganization in the brainstem is not completely expressed in somatosensory cortex, Proc. Natl. Acad. Sci. USA 92 Ž1995. 4264–4268. w11x S.K. Leong, C.K. Tan, Central projections of rat sciatic nerve fibres as revealed by Ricinus communis agglutinin and horseradish peroxidase tracers, J. Anat. 154 Ž1987. 15–26. w12x M.M. Merzenich, J.H. Kaas, J. Wall, R.J. Nelson, M. Sur, D. Felleman, Topographic reorganization of somatosensory cortical areas 3b and 1 in adult monkeys following restricted deafferentation, Neuroscience 8 Ž1983. 33–55. w13x M.M. Merzenich, J.H. Kaas, J.T. Wall, M. Sur, R.J. Nelson, D.J. Felleman, Progression of change following median nerve section in the cortical representation of the hand in areas 3b and 1 in adult owl and squirrel monkeys, Neuroscience 10 Ž1983. 639–665. w14x M.M. Merzenich, R.J. Nelson, M.P. Stryker, M.S. Cynader, A. Schoppmann, J.M. Zook, Somatosensory cortical map changes following digit amputation in adult monkeys, J. Comp. Neurol. 224 Ž1984. 591–605. w15x M.-M. Mesulam, Tetramethyl benzidine for horseradish peroxidase neurohistochemistry: a non-carcinogenic blue reaction product with superior sensitivity for visualizing neural afferents and efferents, J. Histochem. Cytochem. 26 Ž1978. 106–117. w16x G. Paxinos, C. Watson, The Rat Brain in Stereotaxic Coordinates, 2nd ed., Academic Press, San Diego, CA, 1986. w17x T.P. Pons, P.E. Garraghty, A.K. Ommaya, J.H. Kaas, E. Taub, M. Mishkin, Massive reorganization of the primary somatosensory cortex after peripheral sensory deafferentation, Science 252 Ž1991. 1857–1860. w18x E. Rausell, C.G. Cusick, E. Taub, E.G. Jones, Chronic deafferentation in monkeys differentially affects nociceptive and nonnociceptive pathways distinguished by specific calcium-binding proteins and down-regulates g-aminobutyric acid type A receptors at thalamic levels, Proc. Natl. Acad. Sci. USA 89 Ž1992. 2571–2575. w19x E. Rausell, E.G. Jones, Extent of intracortical arborization of thalamocortical axons as a determinant of representational plasticity in monkey somatic sensory cortex, J. Neurosci. 15 Ž1995. 4270–4288. w20x C. Rivero-Melian, ´ J. Arvidsson, Brain stem projections of rat lumbar dorsal root ganglia studied with choleragenoid conjugated horseradish peroxidase, Exp. Brain Res. 91 Ž1992. 12–20. w21x B. Robertson, G. Grant, A comparison between wheat germ agglutinin- and choleragenoid-horseradish peroxidase as anterogradely transported markers in central branches of primary sensory neurones in the rat with some observations in the cat, Neuroscience 14 Ž1985. 895–905. w22x C.E. Schroeder, S. Seto, J.C. Arrezzo, P.E. Garraghty, Electrophysiologic evidence for overlapping dominant and latent inputs to somatosensory cortex in squirrel monkeys, J. Neurophysiol. 74 Ž1995. 722–732. w23x C.E. Schroeder, S. Seto, P.E. Garraghty, Emergence of radial nerve dominance in ‘median nerve cortex’ after median nerve transection in an adult squirrel monkey, J. Neurophysiol. 77 Ž1997. 522–526. w24x E. Taub, Somatosensory deafferentation research with monkeys: implications for rehabilitation medicine, in: L.P. Ince ŽEd.., Behavioral Psychology in Rehabilitation Medicine: Clinical Applications, Williams and Wilkins, Baltimore, MD, 1980, pp. 371–401.