Effects of predegeneration on nerve regeneration through silicone Y-chambers

Locomotor Behavior Analysis in Spinal Cord Injured Macaca radiata after Predegenerated Peripheral Nerve Grafting-A Preliminary Evidence

INTRODUCTION

Primate animal models are being utilized to explore novel therapies for spinal cord injuries. This study aimed to evaluate the efficiency of the transplantation of predegenerated nerve segments in unilateral spinal cord-hemisected bonnet monkeys' (Macaca radiata) locomotor functions using the complex runways.

MATERIALS AND METHODS

The bonnet monkeys were initially trained to walk in a bipedal motion on grid and staircase runways. In one group of trained monkeys, surgical hemisection was made in the spinal cord at the T12-L1 level. In the other group, hemisection was induced in the spinal cord, and the ulnar nerve was also transected at the same time (transplant group). After one week, the hemisected cavity was reopened and implanted with predegenerated ulnar nerve segments obtained from the same animal of the transplant group.

RESULTS

All the operated monkeys showed significant deficits in locomotion on runways at the early postoperative period. The walking ability of operated monkeys was found to be gradually improved, and they recovered nearer to preoperative level at the fourth postoperative month, and there were no marked differences.

CONCLUSION

The results demonstrate that there were no significant improvements in the locomotion of monkeys on runways after the delayed grafting of nerve segments until one year later. The failure of the predegenerated nerve graft as a possible therapeutic strategy to improve the locomotion of monkeys may be due to a number of factors set in motion by trauma, which could possibly prevent the qualities of regeneration. The exact reason for this ineffectiveness of predegenerated nerve segments and their underlying mechanism is not known.

Muscle Reinnervation with Delayed or Immediate Transplant of Embryonic Ventral Spinal Cord Cells into Adult Rat Peripheral Nerve

Muscle denervation is common in various neuromuscular diseases and after trauma. It induces skeletal muscle atrophy. Only muscle reinnervation leads to functional recovery. In previous studies, denervated adult rat muscles were rescued by transplantation of embryonic day 14–15 (E14–15) ventral spinal cord cells into a nearby peripheral nerve. In the present study, changes were made in the environment into which the cells were placed to test whether reinnervation was improved by: 1) prior nerve degeneration, induced by sciatic nerve transection 1 week before cell transplantation; 2) transplantation of 1 million versus 5 million cells; 3) addition of nerve growth factor (NGF) to the transplant. Ten weeks after cell transplantation, axons had grown from all of the transplants. The numbers of myelinated axons that regenerated into the tibial, medial (MG), and lateral gastrocnemius-soleus (LGS) nerves were similar across treatments. The mean diameters of large LGS axons (>6 μm) were significantly larger with nerve degeneration before transplantation. The mean diameters of MG and LGS axons were significantly larger with transplantation of 1 million versus 5 million cells. Silver-stained experimental and control lateral gastronemius (LG) muscles showed axons that terminated at motor end plates. Nodal and terminal sprouts were more common in reinnervated muscles (45–63% of all end plates) than in control muscles (10%). Electrical stimulation of the transplants induced weak contractions in 39 of 47 MG muscles (83%) and 33 of 46 LG muscles (72%) but at higher voltages than needed to excite control muscles. The threshold for MG contraction was lower with transplantation of 1 million cells, while LG thresholds were lower without NGF. The cross-sectional area of whole LG muscles was significantly larger with cell transplantation (immediate or delayed) than with media alone, but all of these muscle areas were reduced significantly compared with control muscle areas. These data suggest that delayed transplantation of fewer cells without NGF assists regeneration of larger diameter axons and prevents some muscle atrophy.

Comparison of different biogenic matrices seeded with cultured Schwann cells for bridging peripheral nerve defects

Tissue-engineering as laboratory based alternative to human autografts and allografts provides "custom made organs" cultured from patient's material. To overcome the limited donor nerve availability different biologic nerve grafts were engineered in a rat sciatic nerve model: cultured isogenic Schwann cells were implanted into acellular autologous matrices: veins, muscles, nerves, and epineurium tubes. Autologous nerve grafts, and the respective biogenic material without Schwann cells served as control. After 6 weeks regeneration was assessed clinically, histologically and morphometrically. The PCR analysis showed that the implanted Schwann cells remain within all the grafts. A good regeneration was noted in the muscle-Schwann cell-group, while regeneration quality in the other groups (with or without Schwann cells) was impaired. The muscle-Schwann cell graft showed a systematic and organized regeneration including a proper orientation of regenerated fibers. All venous and epineurium grafts had a more disorganized regeneration. Seemingly, the lack of endoneural tube like structures in vein grafts lead to impaired regeneration. And, apparently, the beneficial effects of implanted Schwann cells into a large luminal structure can only be demonstrated to a limited extent if endoneural like structures are lacking. A tube offers less area for Schwann cell adhesion and it is more likely to collapse. This underlines the role of the basal lamina, or at least an inner structure acting as scaffold in axonal regeneration. Although the conventional nerve graft remains the gold standard, the implantation of Schwann cells into an acellular muscle provides a biogenic graft with basal lamina tubes as pathway for regenerating axons and the positive effects of Schwann cells producing neurotrophic and neurotropic factors, and thus, supporting axonal regeneration.

Schwann cell metabolic activity in various short-term holding conditions: implications for improved nerve graft viability

Strategies for improvement of nerve regeneration and optimal conditions to prevent Schwann cell (SC) loss within a nerve transplant procedure are critical. The purpose of this study was to examine SC viability, which plays an important role in peripheral nerve regeneration, under various incubation conditions up to three hours. To address this issue, Schwann cell metabolic activity was determined using different independent test methods. The following experimental conditions were compared: SCs prepared from nerves were incubated in (1) isotonic saline solution (2) Dulbecco's modified Eagles medium as used for cell culturing, (3) Hannover bioreactor medium, and (4) Leibovitz's medium. SC metabolic activity of excised rat sciatic nerve was determined at 4°C, 18°C, and 37°C over 3 hrs. The results indicate that SC activity was optimized by the usage of Leibovitz's medium or HBRM at 37°C. Greater SC viability at the time of surgical nerve grafting could contribute to improved axonal regeneration and remyelination after nerve transplantation, and thus more successful functional recovery.

Re-activation of atrophic motor Schwann cells after hypoglossal-facial nerve anastomosis

Facial nerve lesions are common in humans and often require surgical intervention. If repair is delayed, reinnervation can be facilitated by transposing the freshly cut hypoglossal nerve end-to-end directly to the distal facial nerve, allowing for uncompromised hypoglossal axons to reinnervate the denervated facial musculature (hypoglossal-facial anastomosis, HFA). Schwann cells (SCs) in the distal nerve stump have an important function in promoting axonal regeneration by expressing multiple regeneration-associated proteins. Chronically denervated SCs cease to express those factors, but it is unknown whether they can be reactivated by fresh axonal sprouts and regain part of their function. We evaluated SC function and viability in distal facial nerve stump of rats at various time points after chronic denervation as well as following immediate or delayed HFA by assessing their expression of growth-associated protein 43 kDa (GAP-43) and the neuregulin receptors erbB2 and erbB4. Our results show that maximal upregulation of those factors in denervated SCs occurred a few weeks after nerve transection, indicating that a short period of denervation might even be beneficial before nerve repair. Motor SCs denervated for 32 weeks had downregulated their activity and ceased to express the regeneration-associated factors. SCs immediately re-expressed GAP-43, erbB2, and erbB4 following contact with fresh hypoglossal motor axons, demonstrating they are competent to promote regeneration even after long-term denervation.

Effect of end-to-side repair of proximal nerve stumps of transected peripheral nerves on the development of neuroma (experimental study)

OBJECTIVE

Neuroma is a psychologically and physically disabling problematic condition without any current standard therapy. For that reason, we investigated whether end-to-side anastomosis of the proximal end of the transected nerve into the adjacent nerve will prevent the development of neuroma in different types of nerve injuries.

STUDY DESIGN

In this study, hind legs of 18 Sprague-Dawley female rats were used. Six groups were formed. In group I, peroneal nerves were transected and its proximal end was attached end-to-side through the epineural window to the adjacent tibial nerve. In group II, contrary to group I, an epineural window was created in the tibial nerve and the same number of sutures were employed. In group III, tibial nerve was transected proximal to the end-to-side repair site, whereas in group IV, distal segment of the nerve was cut, and an end-to-end repair procedure was repeated. In group V, unlike group I, an approximately 1-cm segment was resected and removed distal (from tibial nerve) to the end-to-side repair site. In group VI, an epineural window was created in the tibial nerve and the same number of sutures were used, and also a 1-cm distal nerve segment was resected. The rats were followed for 2 months, and then all of the groups were evaluated histopathologically, and weights of the posterior muscle groups of hind legs were evaluated.

FINDINGS AND CONCLUSIONS

No neuroma formation was observed in the proximal stumps of peroneal nerve segments in end-to-side repair sites in groups I, III, IV, and V, and proximal stumps of the tibial nerve in group V. In group VI, neuroma formation was observed in the proximal end of the tibial nerve. When weights of the posterior muscle groups of hind legs in groups I and II were comparatively assessed, statistically significant difference was not detected. In conclusion, based on histological data obtained for proximal nerve ends and segments distal to the end-to-side repair sites, we think that end to side neurorrhaphy of the proximal end of the damaged nerve to adjacent nerve will prevent the development of neuroma without injuring the intact nerve segment.

[Paralyzed face. Ansa-cervicalis-nervi-hypoglossi]

INTRODUCTION

For 100 years hypoglossal-facial nerve anastomosis (HFA) has been a common surgical procedure for reanimation of paralyzed mimic muscles of the face after axotomy of the facial nerve. However, the denervation and subsequent scarred degeneration of the target muscles of the hypoglossal nerve often results in unfavorable late effects for speech and swallowing. Therefore, the ansa cervicalis nervi hypoglossi-facial nerve anastomosis (ACHFA) can be an alternative to avoid such late effects. As a branch of the hypoglossal nerve the ansa cervicalis innervates the infrahyoidal muscles. Neck dissection surgery proved that resection of the ansa cervicalis causes no side effects for swallowing because of several nerve anastomoses to the cervical plexus.

PATIENTS AND METHOD

We compared our clinical results of eight cases following a delayed ACHFA with our own experiences after HFA and results from the literature.

RESULTS

We found a reanimation rate lower than usually seen after HFA only in the target muscles of the forehead. This may be caused by a reduced neuronal plasticity of the ansa cervicalis. However, in the target muscles of the other two facial nerve branches we observed the same good results one would expect after HFA. There were no late side effects for swallowing and speech though.

Antibodies to myelin-associated glycoprotein accelerate preferential motor reinnervation

Predegeneration of nerve enhances its ability to support axon regeneration. Trophic factors are upregulated by reactive Schwann cells while potentially inhibitory molecules are removed. These experiments isolate the effects of one such inhibitory molecule, the myelin-associated glycoprotein (MAG), to determine its role in modifying regeneration after nerve repair. Suture of the mouse femoral nerve was followed by daily intraperitoneal injection of antibodies to MAG, antibodies to HNK-1, a specific muscle pathway marker, or no further treatment. Regeneration was assayed by double-labeling the femoral cutaneous and muscle branches with horseradish peroxidase and fluoro-gold after 4 weeks or 6 weeks of regeneration. Four weeks after nerve repair, selective reinnervation of the muscle branch by motoneurons, or preferential motor reinnervation (PMR), was not seen in either controls or L2-antibody-treated animals. In contrast, treatment with MAG antibodies resulted in dramatic PMR. By 6 weeks, the controls had achieved borderline specificity, substantial PMR developed in the L2 antibody group and the MAG group changed little. Blocking access to MAG in the distal nerve stump thus accelerated and enhanced PMR. Sensory regeneration was depressed by both antibody treatments at 4 weeks but recovered by 6 weeks. Antibody administration has a generalized effect on sensory regeneration that is unrelated to the behavior of motoneurons in the same nerve.

Reconstruction of peripheral nerves using acellular nerve grafts with implanted cultured Schwann cells

The bridging of nerve gaps is still one of the major problems in peripheral nerve surgery. The present experiment describes our attempt to engineer different biologic nerve grafts in a rat sciatic nerve model: cultured isogenic Schwann cells were implanted into 2-cm autologous acellular nerve grafts or autologous predegenerated nerve grafts. Autologous nerve grafts and predegenerated or acellular nerve grafts without implanted Schwann cells served as controls. The regenerated nerves were assessed histologically and morphometrically after 6 weeks. Predegenerated grafts showed results superior in regard to axon count and histologic appearance in comparison to standard grafts and acellular grafts. The acellular nerve grafts showed the worst histologic picture, but axon counts were in the range of standard grafts. The implantation of Schwann cells did not yield significant improvements in any group. In conclusion, the status of activation of Schwann cells and the stadium of Wallerian degeneration in a nerve graft might be key factors for regeneration, rather than total number of Schwann cells. Predegenerated nerve grafts are therefore superior to standard grafts in the rat model. Acellular grafts are able to bridge nerve gaps of up to 2 cm in the rat model, but even the addition of cultivated Schwann cells did not lead to results as good as in the group with autologous nerve grafts.

Sorting of Regenerating Rat Sciatic Nerve Fibers with Target-Derived Molecules

The functional outcome of microsurgical repair of divided nerves is disappointing since many regenerating axons fail to reach appropriate targets. Sorting of regenerating axons according to target tissue might be used to improve functional regeneration. The aim of the present study is to see if regenerating axons can be sorted into functionally different bundles with target-derived molecules. The proximal stump of the adult rat sciatic nerve was sutured into the inlet of a silicon Y-tube. The two branches of the Y-tube were filled with agarose primed with filtrates prepared from skin and muscle homogenates from the operated rat. The tibial and sural nerves were inserted in the two branches of the Y-tube. Six weeks later the sciatic nerve axons showed vigorous regeneration into both branches. Electron microscopic examination of regenerated nerve segments showed numerous myelinated and unmyelinated axons. The proportion of myelinated axons was significantly larger in the muscle-gel branch than in the skin-gel branch. Retrograde tracing from the nerve regenerates with Fast Blue and Fluoro-Ruby showed that ventral horn neurons at L4-L5 segmental levels were preferentially labeled from the muscle-gel branch. Neurons in corresponding dorsal root ganglia were labeled from both Y-tube branches (no significant numerical difference). A few neurons of both types contained both tracers. Measurements revealed that sensory neurons labeled from the muscle-gel branch were significantly larger (mean perikaryal area 870 microm(2)) than neurons labeled from the skin-gel branch (mean area 580 microm(2)). We conclude that regenerating motor and sensory axons can be sorted with target-derived molecules.

A role of migratory Schwann cells in a conditioning effect of peripheral nerve regeneration

The common peroneal nerve in mice was conditioned by axotomy around the head of the fibula. At various intervals from 1 day to 2, 3, 5, 15, and 25 days, a test lesion was made by axotomy 15 mm proximal to the conditioning lesion site. The proximal stump of the transected nerve was sandwiched between two sheets of thin plastic film and remained in vivo for various intervals from 3 h to 6, 9, 12, 24, 48, 72 and 96 h. The regenerating axons were visualized on the film with silver nitrate impregnation. Schwann cells were visualized migrating onto the film using immunohistochemistry with anti-S-100. To determine the effects of migratory Schwann cells on axonal outgrowth, a film model was established on one limb. After the nerve stump was removed from the film, the treated film was transferred to a new lesion on the contralateral limb and 2 days later the film was harvested for histological examination. Conditioned by a prior axotomy more than 3 days earlier, regenerating axons sprouted within less than 1 h after the test lesion was established and grew naked at five times higher rate: The growth rate was similar to that observed during regeneration in the presence of migratory Schwann cells (ordinary type). After a short interval, the axons, which had been ensheathed by migratory Schwann cells (reactive type), continued growing at a significantly (P < 0.01) higher rate. The reactive type of cells had fewer numbers of branches and higher activity in promoting axonal outgrowth than the ordinary type. Thus, both ordinary and reactive types of cells played key roles in initiating and maintaining a conditioning effect, respectively.

Contributions of pathway and neuron to preferential motor reinnervation

Motor axons regenerating after transection of mixed nerve preferentially reinnervate distal muscle branches, a process termed preferential motor reinnervation (PMR). Motor axon collaterals appear to enter both cutaneous and muscle Schwann cell tubes on a random basis. Double-labeling studies suggest that PMR is generated by pruning collaterals from cutaneous pathways while maintaining those in motor pathways (the "pruning hypothesis"). If all collaterals projecting to muscle are saved, then stimulation of regenerative sprouting should increase specificity by increasing the number of motoneurons with at least one collateral in a muscle pathway. In the current experiments, collateral sprouting is stimulated by crushing the nerve proximal to the repair site before suture, a maneuver that also conditions the neuron and predegenerates the distal pathway. Control experiments are performed to separate these effects from those of collateral generation. Experiments were performed on the rat femoral nerve and evaluated by exposing its terminal cutaneous and muscle branches to HRP or Fluoro-Gold. Crush proximal to the repair site increased motor axon collaterals at least fivefold and significantly increased the percentage of correctly projecting motoneurons, consistent with the pruning hypothesis. Conditioning the nerve with distal crushes before repair had no effect on specificity. A graft model was used to separate the effects of collateral generation and distal stump predegeneration. Previous crush of the proximal femoral nerve significantly increased the specificity of fresh graft reinnervation. Stimulation of regenerative collateral sprouting thus increased PMR, confirming the pruning hypothesis. However, this effect was overshadowed by the dramatic specificity with which predegenerated grafts were reinnervated by fresh uncrushed proximal axons. These unexpected effects of predegeneration on specificity could involve a variety of possible mechanisms and warrant further study because of their mechanistic and clinical implications.

Selective inhibition of early axonal regeneration by myelin-associated glycoprotein

When the distal stump of a transected peripheral nerve is brought into the vicinity of the proximal nerve stump, the regenerating axons advance toward it across the gap. Similar results are obtained when a predegenerated nerve segment is used. However, when a nerve segment subjected to proximal axotomy 7 days earlier (7-day nerve segment) was placed close to the proximal end of a freshly cut nerve at a distance of less than 1.5 mm, there were neither regenerating axons nor sprouts. The same inhibition of axonal regeneration was also exhibited when a nerve segment subjected to axotomy 9 to 14 days earlier was used. To examine the inhibitory effect of the nerve segments on established regenerating axons, we positioned a 7-day nerve segment in close apposition to a proximal nerve end at 2 or 3 days after transection. The growth of the 3-day-old regenerating axons, already ensheathed by Schwann cells, was not disturbed, but the 2-day-old regenerating axons, consisting of naked axons, were eliminated by the 7-day nerve segment. It is assumed that the findings reflect a mechanism serving to eliminate abundant sprouts and immature axons, probably conferring optimum regeneration and maturation of outgrowing pioneer axons. The inhibitory effect on abundant sprouts and immature axons was completely blocked by local application of antibodies to myelin-associated glycoprotein (MAG). The MAG-containing cells appeared at 6 to 12 days after axotomy.

Predegenerated nerve allografts versus fresh nerve allografts in nerve repair

This study reevaluated the possibility of using predegenerated nerves as donor nerve allografts for nerve repair and compared the results of functional recovery to those obtained after standard, fresh nerve allograft repair. Twenty donor rats underwent a ligature/ section of the left sciatic nerve 4 weeks before nerve graft harvesting. Forty recipient rats underwent severing of the left sciatic nerve leaving a 15-mm gap between the nerve stumps. Graft repair was undertaken using either the predegenerated left sciatic nerve of the 20 donor rats (predegenerated group, 20 recipient rats) or the normal right sciatic nerve of the 20 donor rats (fresh group, 20 recipient rats). Recovery of function was assessed by gait analysis, electrophysiologic testing and histologic studies. Walking tracks measurements at 2 and 3 months, electromyography parameters at 2 and 3 months, peroperative nerve conduction velocity and nerve action potential amplitude measurements at 3 months, as well as assessments of myelinated nerve fiber density and surface of myelination showed that fresh and predegenerated nerve grafts induced a comparable return of function although there was some trend in higher electrophysiologic values in the predegenerated group. The only slight but significant difference was a larger mean nerve fiber diameter in the nerve segment distal to a predegenerated nerve graft compared to a fresh nerve graft. Although our study does not show a dramatic long-term advantage for predegenerated nerve grafts compared to fresh nerve grafts, their use as prosthetic material is encouraging.

Delayed hypoglossal-facial nerve suture after predegeneration of the peripheral facial nerve stump improves the innervation of mimetic musculature by hypoglossal motoneurons

Surgical reconstruction of the facial nerve is common clinical practice following destruction of the intracranial facial nerve. Delayed hypoglossal-facial anastomosis (HFA) is the procedure of choice, although the effect of delay on outcome remains unclear. To study the effect of delayed anastomosis on reinnervation, we sutured the proximal stump of a freshly transected hypoglossal nerve of Wistar rats to the distal stump of the ipsilateral facial nerve, which had been transected 7-56 days earlier. Animals that had received HFA without delay served as the control group. Forty days after HFA, horseradish peroxidase (HRP) was injected into the whisker pad; 2 days later, the animals were killed. Reinnervation was assessed by determining the proportion of labeled neuronal cell bodies in the brainstem. The control group had 68% reinnervation of these muscles by hypoglossal neurons and had 32% reinnervation by facial neurons. When the distal facial nerve had been allowed to degenerate for 7 days before HFA, reinnervation of the hypoglossal nerve decreased to 54%, and reinnervation by the facial nerve increased to 46%. However, after a delay of 10-56 days, the hypoglossal fraction increased and stabilized at 77%, and the facial motoneuron fraction decreased to 23%. The presence of new neuromuscular junctions was confirmed by HRP labeling of motor end plates in vivo and by electromyography. We conclude that, under the conditions of hypoglossal-facial crossed nerve suture, the predegeneration of the distal stump of a transected facial nerve enhances the reinnervation of facial muscles by hypoglossal axonal sprouts.

Peripheral nerve regeneration through short- and long-term degenerated nerve transplants

The present study was designed to compare regenerative potential of normal and degenerated nerve grafts. Peripheral nerves in rats were induced to undergo in situ degeneration for a period of 6 weeks, 3, 6 and 12 months. During early phase of denervation the myelin and axons degenerated and were absorbed. With prolonged denervation (i.e. 12 months), such nerves were reduced in size and exhibited extensive fibrosis. A 2 cm long segment of the degenerated nerve was transplanted in an surgically created gap in the host peroneal nerve to evaluate their regeneration supporting ability. Regeneration of host axons occurred rapidly through nerves degenerated for a period up to 3 months. The extent of regeneration was compromised in 6-month degenerated nerve group, and was significantly reduced in the 12-month degenerated nerve grafts. These results show that with extended degeneration interval, the regeneration supporting ability of nerves is compromised. It is concluded that nerve repair should not be excessively delayed in order to compromise recovery.

Predegeneration enhances regeneration into acellular nerve grafts

In the present study, we determined the regeneration rate and the initial delay in rat sciatic nerve grafts first made hypercellular by predegeneration then acellular by freeze-thawing. 7-day predegenerated nerve pieces from the distal nerve stump on the right side were made acellular by repeated freeze-thawing and inserted as grafts into a 10-mm long freshly created defect on the left contralateral side. Freshly made (no predegeneration period) acellular nerve grafts were used as controls. Both types of grafts supported outgrowth of regenerating axons as demonstrated by the sensory pinch test. However, the predegenerated acellular nerve grafts had a significantly shorter initial delay period (2.7 days) as compared with freshly made acellular nerve grafts (9.5 days). The initial delay period for predegenerated acellular nerve grafts was similar to that for fresh cellular nerve grafts but significantly longer than that for predegenerated cellular nerve grafts [24]. The rate of regeneration appeared independent of the type of grafts used. We suggest that modifications of the basal lamina and/or factors produced during the predegeneration period by non-neuronal cells survive the freeze-thawing cycle and account for the decrease in the initial delay period.

Perspectives on axonal regeneration in the mammalian CNS

In the CNS of mammals, axonal regeneration is limited by inhibitory influences of the glial and extracellular environment. Myelin-associated inhibitors of neurite growth, as well as some properties of so-called 'reactive astrocytes' which make the environment non-permissive for axonal growth, contribute to the inhibitory nature of the mammalian CNS. Furthermore, mechanisms for effective removal or neutralization of inhibitory components of cell debris are lacking in the mature mammalian CNS. However, in a permissive environment, mammalian CNS axons are able to regrow, to recognize target areas and to re-establish functional synapses with target neurones. Moreover, recent observations suggest that guiding molecules, like those required for axon guidance in the developing CNS, become expressed after lesions. Regenerating CNS axons seem to be able to recognize such guidance cues. Thus, even regenerating CNS axons of mammals might ultimately succeed in re-establishing topographically ordered functional synapses in their target regions.

Can sensory and motor collateral sprouting be induced from intact peripheral nerve by end-to-side anastomosis?

The possibility that collateral sprouting could occur from intact axons in an undamaged sciatic nerve was studied in the rat by suturing either a 7-day predegenerated or a fresh nerve segment in an end-to-side fashion to the sciatic nerve proper. Following a 14- or 35-day recovery period, the pinch reflex test was performed on the transplanted segment to demonstrate the presence of sensory axons. The majority of cases, using a predegenerated nerve segment but not a fresh segment, responded positively. Neurofilament staining and histological examination confirmed the presence of axons in the attached nerve segment. In another series of experiments, the proximal peroneal fascicle was ligated and cut. Following a 7-day predegeneration period the distal stump was sutured end-to-side to the ipsilateral tibial fascicle. After 90 days, stimulation of the tibial nerve proximal to the attached site induced substantial contraction in both the native gastrocnemius muscle and the foreign tibialis anterior muscle. These findings suggest that collateral sprouting may occur from intact axons, perhaps induced by factors emanating from the attached nerve segment, and subsequently make functional peripheral connections.